,1 V r^^-: -> •• r ■.';?t^ • ^ ^«- .<-r^- "^- V- ♦ .■-^? A -^ V * ' weeks old. teria, and in future all stock cultures intended for distribution by the Office of Soil-Bacteriology and Plant-Xutrition Investigations to farmers and other experimenters in the United States will be tested in this manner. tCir. 120] rCir. 120— B.] THE WORK OF THE SAN ANTONIO EXPERIMENT FARM IN 1912/ By S. H. Hastings. Farm Superintendent, Office of Western Irrifjation Agri- (uttiire. INTRODUCTION. The work of the Snii Antonio Experiment Farm is devoted to the investigation of agricnltiiral problems peculiar to large areas in the fcouthwestern United States, where the conditions of soil and climate are similar to those at San Antonio. The more important lines of work carried on are tillage and rotation experiments; breeding, variety testing, and dilTerent planting methods for cotton; variety Fio. 1. — View sbovviui? the buildiufjjs on the S:ui Antouio Expcriiuent Farm. testing of grain sorghum and broom corn; corn breeding and variety testing; a few forage-crop experiments; variety testing of peaches, jDlums, apricots, persimmons, grapes, walnuts, almonds, and other fruits, including Chinese dates. In addition to such horticultural work as is mentioned above, much attention is being given to finding stocks better adapted to the local conditions than those now used. 1 Issued Apr. 5, 101.",. The San Antonio Experiment Farm comprises about 125 acres of hind situated about (> miles south of San Antonio, Tex. The tract belongs to the city of San Antonio and is leased to the Department of Agriculture. Al)out 80 acres of the land are under cultiva- tion, and of these are irrigated. The farm is equipped with the buildings (fig. 1 I nec- essary for storage, laboratory, and office purposes and for employees' quarters. The e.xperimcnt farm is under the direction of the Office of Western Irrigation .Vgriculture of the Bureau of Plant Industry and is maintained from the funds of the Department of .\griculture. Previous general reports on the work of the farm were published in 10(»S and moo as Bureau of Plant Industry Circulars Nos. l.T and ?A. [Cir. 120] ' 8 CIRCULAE NO. 120, BUREAU OF PLANT INDUSTRY. The testing of trees and shrubs for ornamental purposes is being emphasized. The location of the experiments in 1912 is shown in figure 2. COOPERATIVE WORK, The following are the offices of the Bureau of Plant Industry co- operating in the work at San Antonio, the approximate area occupied by each being noted : The Office of Ac- climatization and Adaptation of Crop Plants and Cotton- Breeding Investiga- tions has about 11 acres devoted to corn and cotton. The Office of Foreign Seed and Plant In- troduction has ap- proximately 1^ acres devoted to the test- ing of new plant in- troductions. The Office of Crop Phys- iology and Breeding Investigations has about 2 acres de- voted to testing the Texas Avild peach [Prunus texana) and its hybrids and to fig and pistache ex- p e r i m e n t s. Ap- proximately 3 acres are used by the Office of Corn In- vestigations in test- ing corn varieties. About li acres are used by the Office of Alkali and Drought Kesistant Plant Investigations in its work with pomegranates and olives. The Office of Cereal Investigations cooperates in the variety testing of grain sorghum and broom corn. CLIMATIC CONDITIONS. The first part of the season of 1912 was very favorable to crop arowth. and as a result the yields of early crops, such as corn, oats. [Cir. 120] Pig. 2. Diagram of the San Antonio Experiment Farm, showing the arrangement of the fields and the location of the experiments in l'.n2. WORK OF THE SAN ANTONIO EXPERIMENT FARM IN 1912. 9 and sorghum, were good. The preceding winter was one of unusual severity and the amount of winter rainfall was somewhat above the normal. From June 23 until late in September practically no rain fell, the total precipitation for the months of July and August being only 0.33 inch. This continued drought of nearly three months cut the cotton crop short. The spring was late and consequently the corn and other early-planted crops were somewhat late in maturing. The total precipitation for the year, as measured at the farm, was 26.37 inches,^ which is practically the normal for this section, but somewhat higher than the mean annual rainfall for the period 1907 to 1912, inclusive. The meteorological observations at the experiment farm are made in cooperation with the Biophysical Laboratory of the Bureau of Plant Industry. Table I shows the summaries for 1912 compared with the means for the six-year period 1907 to 1912, inclusive. Table I. — Summary of meteorological observations at the San Antonio Experi- ment Farm, 1D07 to 1912, inclusive. Precipitation (Inches). Item. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. Total. Average, 6 years, 1907 to 1912, in- clusive 0.54 .31 2.23 6.21 1.66 2.30 2.96 2.04 2.47 1.64 1.28 3.42 1.54 .08 1.63 .25 1.15 1.53 2.60 2.92 2.52 1.76 1.S8 3.91 22.46 For the year 1912... 26.37 Evaporation (Inches).^ Average, 6 years, 1907 to 1912, in- elusive 2.72 2.35 3.12 3.35 4.68 3.05 5.32 3.88 6.63 7.39 8.40 7.02 8.69 10.59 9.39 10.65 7.31 8.52 5.23 5.41 3.21 3.23 2.38 1.83 67.07 For the year 1912... 67.26 Daily Wind Velocity (Miles per Hour).^ 1911, highest.. 1912, highest., 1911, lowest.. 1912, lowest.. 1911, average. 1912, average. 10.4 8.2 1.12 .77 5.60 3.16 15.9 11.2 1.45 .62 7.02 4.00 9.0 6.4 2.24 .86 5.20 3.50 10.6 6.0 3.28 .84 6.10 2.68 9.2 5.9 2.17 1.22 5.60 3.23 11.9 8.4 2.34 1.08 6.40 2.92 12.5 7.3 2.82 1.38 6.86 4.27 12.1 7.6 1.82 2.30 4.59 4.83 6.6 6.5 1.98 1.42 3.91 3.75 8.5 7.6 .86 .89 3.76 3.84 9.5 5.7 1.06 .76 3.24 2.68 8.0 .97 .81 3.36 2.80 1 The rainfall in the city of San Antonio in 1912, as reported by the United States Weather Bureau, was 23.7 inches, which was 3.2 inches below the normal. 2 The evaporation measurements missing are as follows: 1909 —April, 2 davs; June, 3 davs; July, 6 days; August, 1 day; TMoveniber, 1 day 1910.— January, 7'days; February, 2 days; June, 3 days; July, 5 days; October, 4 days; December, 3 days. 1911.— April, 1 day; July, 2 days. 1912.— February, 2 days; April, 1 day; December, 2 days. » Wind velocities are reported for the years 1911 and 1912 only. 84899°— Cir. 120—13 2 10 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. Table L — Xuiiiiikuii of iii(l(<. 8li. 5 82 108.0 Absolute m a x i- mum, 1912 84.0 7.S. 5 S2.0 90. lo:i.o 100. 5 105.0 104.0 101.0 91.0 Sl.O /! 105. Absolute minimum, 6 vears, 1907 to 1912, iticlusive 12.0 i;i.() lil.O ;is. :i9.o 56. 64.0 64.0 11.0 :i2.o 15.0 1/ 12.0 Absolute minimum. 1912 16.5 16.0 :!4.0 ys. 47.0 59. 66.0 68.0 54.5 51.0 25. 25 1 (i, Mean, 6 years, 1907 to 1912, inclusive.. .'■)4.1 54.7 65.0 6X. 6 75.2 82. 7 85. 2 85.3 80.3 69.6 59. 6 :-M 69.2 Mean, 1912 46.6 49.9 55.4 67.4 76.5 77.8 85. 1 813. 8 82.0 71.8 59.4 48.6 67.3 KiLLlNi; Fro.sts. I^ast in spring. First in autumn. Length of frost- "i ears. Mini- Mmi- Date. mum temper- ature. Date. mum temper- ature. period. "F. "F. Days. 1907 Feb. 8 Feb. 20 Feb 25 29 24 30 26 29 30. 5 Nov. 12 Nov. 14 Dec. 6 ()ct. 29 Nov. 13 Nov. 2 32 29 31 32 31 29.5 ■ 277 1 <»()S 268 1909 284 1910 do "27' 246 1911 ...do 261 1912 Feb 245 ROTATION AND TILLAGE EXPERIMENTS.' The rotation and tillag;e experiments, which are eonihicted on S'2 plats of one-fourth of an acre each, were continued as ]>reviously. except that milo was substituted for corn on four phits and for oats on one phit. The results of the rotation experiments indicate that crop rotation is an important factor in crop production in this section of Texas, the yields of crops grown in a suitable rotation being on the whole uniformly higher than when grown continuously on the same land. The average yields of all crops except cotton were the highest of any 3^ear since the work was started in 1009. Table II giACS the crops in the rotation experiments, tlie number of plats planted to each crop in 1912, the average yields per acre, and the highest and lowest yields per acre in 1912. 1 These experiments are under the direct supervision of Mr. C. U. Letteer, assistant. ICir. 120]' WORK OF THE SAN ANTONIO EXPERIMENT FARM IN 1912. 11 Table II. — Arcr(i(>r ijicUlx prr ncrr nf crops in flic lohitioii cdiKriinciits, Saii Antonio E.rinrinicnf Fniin. I'.) 12. Crop. Corn bushels . Dwarf milo do Oats, grain ' do — Cotton 2 pounds. . Sorghum: 3 4 1 foot drills ton.s. 8-inch drills do Oats, hay ' do Average yield, 1907 to 1911, inclusive. 14.6 6.5 585.3 4.86 2.06 .83 Yield in 1912. Niunbor of plats. 26 5 10 25 5 3 4 .\veragp. 34.1 40.0 26. 75 621.5 4.03 4. 68 2.82 Highest. 42.3 52.0 37.0 818.0 4.28 5.49 .•i09 Lowest. 24.3 32.5 19.1 448.0 3.82 4.16 2. 53 1 No oat yields in 1907 and 1908. 2 geed cotton. 3 Sorghum not planted in S-inch drills in 1908. MANURING. The effect of barnyard manure on crop yields in the rotations Avas more noticeable this year than ever before. The average yield of corn from all plats where manure Avas applied at some time during the course of the rotation was slightly greater than the average of corn plats in corresponding rotations not manured. The same was true of cotton, but manuring decreased the yield of oats for grain very noticeably. The oats on plats which had been manured grew very rank and lodged much more than on plats which had not been manured. The excessive vegetative growth and consequent lodging probably account largely for the decrea.sed yield of grain. While the results indicate the value of barnyard manure, the dif- ference in favor of manure is much less on crops groAvn in a rotation than where grown continuously on the same land and manured each year. (See figs. 3 and 4.) Table III gives the yields for 11)12 and the average yields for lUlO, 1911, and 1012 of crops gi'own continuously on the same land manured ^ each 3'ear compared with plats not manured. Table III. — Yield - for ][)l.i and nrcnKjc yields for tniO. 1911. and li)12 of croijx on uuinured and niunaniircd pints phintcd continiKiuslij to the sunic crops lit the San Antonio E-rpcriincnf Farm. Manured. Not manured. Difference in fa- ■\'or of manuring. Crop. Average, 1910 to 1912. 1912. .\verage, 1910 to 1912. 1912. 1910 to 1912. 1912. Corn 15.9 30.5 52.0 572.0 11.6 26.6 32.5 474.0 4.3 3 9 Dwarf milo ^ 19.5 Cotton 455. 3 397.3 58.0 98 1 The manure is applied at the rate of about 15 tons per acre. 2 Corn and milo in bushels per acre; cotton in pounds of seed cotton per acre. 3 The plats from which dwarf milo yields are given had previously been planted to corn for three years. [Cir. 120 J 12 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. SUMMER FALLOWING. Three years' results have now been obtained from the tests in summer fallowing land for corn, oats, and cotton. As in 1911, the yields of crops on land summer-fallowed the previous season were generally low, corn yielding at the rate of 24.7 bushels per acre, as compared with the average of 34.1 bushels per acre on the 20 plats in the rotation experiments. The corn on the summer-fallowed plat yielded the lowest, save one, of any of the 26 plats. Cotton on summer-fallowed land yielded at the rate of 448 pounds of seed cotton per acre, as compared with an average of 621.5 pounds per acre on the 25 plats of cotton, giving the lowest yield of the 25 plats. Fig. i!. — Cotton on plat B 5-:!, land continuously cropped and not manured. This plat has yielded an average of 307 pounds of seed cotton per acre during the past three years. Compare with flsuro 4. (I'hotojiraphcd Juno 2G, lOli;. ) Oats for grain on a fallowed plat yielded at the rate of 37 bushels per acre, the highest yield obtained from the 10 plats of oats. The average yield from 10 plats Avas 26.75 bushels per acre. While the oats on summer- fallowed land yielded somewhat higher than the others, the increase was not sufficient to indicate that summer fallow- ing is a desirable practice under the conditions at San Antonio, even for oats. SUBSOILING. Subsoiling tests have been a rather important j)art of the rotation and tillage experiments. The results have been sunnnarized and pub- lished in Circular No. 114 of the Bureau of Plant Industry. [Cir. 120] WORK OF THE SAN ANTONIO EXPERIMENT FARM IN 1912. 13 The results from subsoiling in 1912 were corroborative of those of previous j-ears, namely, that subsoiling does not materially in- crease the yields of crops, and in many instances decreases the yields, and that, owing to its being an expensive operation, it can not be rec- ommended as a regular farm practice in connection with corn, oats, and cotton in the San Antonio region of Texas. ROOT-ROT. Root-rot, a fungous disease of plants,^ is doubtless one of the most serious diseases with which farmers have to contend in the Black Lands of Texas. It affects such crops as cotton, cowpeas, and alfalfa, Fig. 4. — Cotton on plsit B 5-4, land continuously cropped and fertilized each year with harnyard manure. This plat has yielded an average of 4r)r) pounds of seed cotton per acre during the past three years. Compare with figure :'.. (Photographed .Tune 26, 1912.) but does no perceptible damage to plants belonging to the grass fam- ily, such as corn, oats, wheat, etc. In many cotton fields it causes the premature death of a large proportion of the plants. It was observed in 1912 that in a number of the rotations with cotton and corn the root-rot was much more widespread and did more damage to the cot- ton on plats which were si^ring-ploAved than on plats which were summer or fall plowed. The same condition was apparent in 1911. It was also observed in 1912 that where cotton was grown in rota- tion with corn or oats the damage due to root-rot was much less noticeable than on plats continuously planted to cotton. 1 This disease is caused by 0.:uniiiiii uitniironuii. For an account of tlie diseusc and methods of lessening its damage, see the paper entitled " The control of Texas root- rot of cotton," in Bureau of Plant Industry Bulletin 102. [Cir. 120] 14 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. HORTICULTURAL WORK.' THE MEXICAN SEEDLING PEACH. The Mexican seedling peach trees- l)()re a heavy crop of fruit, and the early prospects were good that more data on the comparative merits of different trees might he secured. The earlier ripen- ing trees gave a heavy yield of good fruit, but the continued drought of midsununer dam- aged all the jieaches from the later maturing trees, making the tests with these trees nearly worthless. Nearly all the trees bore an exceptionally heavy crop, but they Avere not able to liold the fruit through such a long period of drought. The trees are becoming over- crowded, owing to their size and close planting, bringing out the point that to obtain the best results the distance be- tween the trees will have to be greater than is ordinarily adopted in more humid sec- tions. The trees were planted 1() feet apart, but it is now ap- parent that better results would be secured if they were 22 feet apart. The Office of Foreign Seed and Plant Introduction put in 324 buds from what were considered the best 9 trees of the ]\Iexican seedling orchard. These budded trees are to be distributed to cooperators in southwest Texas to test their value undei tions. I,'i,;. 5. — English walnut si':iftfcl on nativf Texas black walnut iu February. Uni.'. (PhotoKiaplied August 1!S. I!tl2.) varviu"' condi- 1 The horticultural experiments were under the direct charge of Mr. U. E. Hlair, scien- tific assistant. = The seed from which these trees wer.> produced was collected in Mexicii by Mv. Cilbert Onderdonl<. under the direction of the Officp of Foreign Seed and Plant Introduction. They are listed under S. P. I. Nos. n:!L'0 and U:','21. |fir. lliU] WORK OF THE SAN ANTONIO EXPERIMENT FARM IN liil2. 15 OTHER PEACH VARIETIES. Ill the variety peach orchard on Field Al. where there are 35 varieties of peaches, set out in lOOO, an unusually heavy cro]) of fruit set. P2ven the northern peaches in most instances were heavily loaded for the first time in the history of the orchard. This should not be taken as an indication that this class of peaches is adapted to these conditions. Those belonging to the South Chinese type are the most reliable bearers. Of the varieties on trial the following have proved to be the best: Pallas, Honey, Im- perial, and Triana. The wild peach from China {Anujf/dnhis davidiana) is proving to be excep- tionally good for peach stock. l)ut there is difficulty in securing seed, as thus far it has not fruited here, with the exception of two fruits which ripened in 1912. PLUMS. The tests of plums include 1<) varieties that have been under trial since 190G. The lasts so far con- ducted indicate that the plum is probably the most reliable fruit for the San Antonio section. The American-Japanese hybrids bear somewhat more heavily than the pure Japanese sorts. The plums that have proved the best adapted to the conditions at San Antonio are the Gonzales, Burbank, Wick- son, Terrel, El Paso, and Trans- fi parent. (J. — Xoni:i urnpc m'Mftcd (Ui ii;ili\i' mustang grnpe in February, IHIL'. i I'lio- tographed June -18, 191:.'. i NATIVE TREES AND STTPFP.S. Much work has l)een done on the domestication of nati\e trees and shrubs suitable for use as grafting stock (fig. 5) oi- for crossbreeding. (fir. l:.'ii| 16 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. Hybridization and grafting work has been done on the peach, phim, persimmon, and grape. Successful crosses between the native Primus texana and the peach and phmi have been made. The grafting of the better types of grapes on the native mustang grape has been done successfully (fig. 6). Some attention is also being given to grafting the English walnut on the native black walnut. PERSIMMONS. The prof)agation of persimmons has been tried, a special effort being made to utilize as a stock the native Texas persimmon {Dios- pyros texana), which grows so abundantly and persist entl}' on the semiarid lands of western Texas and in the limestone canyons. This tree has long been regarded as a probable stock for the Japanese persimmon, but until recently it was impossible to produce a union. Successful results have lately been accomplished by inarching with Diosjyyros texana seedlings small enough to be handled in pots. In this method of inarching, the small native seedling is transferred to a paper pot after most of the soil has been removed from the roots, and the pot is filled with sphagnum. It is necessary that the pots be small and light, for under normal summer conditions the wind would prevent securely fastening a heavy pot in position long enough to permit the formation of a union. The stock and scion will com- monly unite in about 30 days, but as the persimmon gi'ows rather slowly even more time is desirable. Until this method was devised, no successful Dlospyros kali-i plants had been produced on I), texana stock. A few crown grafts were made to start, but no others. Budding has been attempted repeatedly by shield, patch, flute, and ring methods, with no success whatever. The following combina- tions have been successfully accomplished here by the inarch method described : D. kaki on D. texana stock. D. virginiana on D. texana stock. D. texana on D. kaki stock. D. texana on D. virginiana stock. A small planting of varieties of Japanese persimmons worked on the native stock is to be made under dry-land conditions as soon as the trees can be propagated. Several have already been set to orchard positions. As most species of the Japanese persimmon suffer froui chlorosis on the soils of the San Antonio region, it is believed that the use of this resistant stock may be a remedv for this trouble unless the scion proves to rapidly outgrow the stock. [Cir. 120] WORK OF THE SAN ANTONIO EXPEEIMENT FARM IN 1912. 17 ORNAMENTAL PLANTS. Ornamental trees and shrubs suitable for the San Antonio section have been tried in large numbers. Particular attention is also being given to native trees and shrubs not already known or whose desir- able qualities are not fully appreciated. In addition to the native trees and shrubs, about 100 varieties of roses and a collection of bam- boos, palms, yuccas, agaves, and many others have been assembled. FORAGE CROPS. The results of all the forage-crop experiments conducted at the farm from 1908 to 1912, inclusive, have been assembled and published as Circular Xo. 106 of the Bureau of Plant Industry. This circular contains an extensive discussion of the results with the forage crops tested during the past five years. For this reason, only brief men- tion will be made here of the main features of the work in 1912. Among the more conspicuous new forage crops that are particu- larly well adapted to the section may be mentioned Canada field peas and Sudan grass. So far Canada field peas have given a higher hay yield than oats, and apparently there is at least one variety that will stand nearly, if not quite, as much cold. Last winter there were on trial three lots, known as S. P. I. Nos. 30307, 18806, and 30134. Dur- ing the winter there was a minimum temperature of 15° F. Xos. 30307 and 18806 were almost completely killed out, but, while the plants on the other planting (No. 30134) were killed nearly to the ground, new growth was put out and a grain yield of 14 bushels per acre was obtained. As a winter cover croj) this is the most promising that has been tested. Excellent results were obtained with Sudan grass in both 1911 and 1912. The yield of Sudan grass in 1912 was 5.66 tons per acre, based on the weights from a one-tenth-acre plat. Sorghum planted in the same way on the rotation fields gave a yield of 4,68 tons per acre. A rate-of-seeding test consisting of seven one-tenth-acre plats of Sumac sorghum in 8-inch drills was conducted, as there is a variance of opinion among farmers as to the best rate, quality of forage and yield considered. The rates varied from 26 to 174 pounds per acre. The highest yield was obtained from the plat seeded at the rate of 88 pounds per acre, or slightly more than 1^ bushels. The quality of the forage was apparently equal to that j)roduced by the heavier seedings and better than that in the thinner seedings. Japanese sugar cane, which has been grown here for the past three seasons, gave a yield in 1912 of 13.08 tons per acre, the average yield for the past three years being 12.84 tons per acre. This crop is grown under irrigation. 84899°— Cir. 120—13 3 18 CIRCULAR XO. 120, BUREAU OF PJ.AXT INDUSTRY. VARIETY TEST OF COTTON.' A variety test of cotton was conducted with seven varieties and five selections of Triumph. Sixteen-rod rows of each variety were planted in triplicate. Table IV gives the results of the test. Table IV. -Vielcls obtained in a variety test of cotton at the San Antonio Experiment Fodh. 1912. Seed cotton per acre. Percentage. \ ariety. Lint. Stand. 1 Pounds. 735 635 580 580 525 525 520 496 480 480 410 335 32.4 35.5 37.6 37.8 34.2 32.8 37.2 27. S 37.9 38.2 35.7 34.0 100 R \ 1 000 fTriiimnhl 90 97 s \ Q90 (■ Trill mnh"! S9 78 89 Trinmrth FPtipral seed -- 93 T)iiraTifrn 87 TriumDh 152 87 Triiimnh MeViane - 94 \cala 3 SO 69 1 Percentage of .stand determined from an actual coimt of the plants in each row. - Lint very short. 3 Lint but little better than Triumph. As is shown, the Vergatus gave the highest yield, but the lint was very short, making it an undesirable variety. GRAIN SORGHUMS. The experiments in grain sorghums reported on last year - were continued. Both variety and time-of-planting tests were made. Table V gives the average yields of the varieties for three plantings. Table V. — Yields obtained in a variety test of grain sorghum at the San Antonio Experiment Farm in 1912. Variety. Bushels per acre. Variety. Bushels per acre. 50.5 46.8 36.5 33.1 29.7 27.9 White sorghum 25.4 Rtatifiarfi milo White durra 22.6 Standard Blackhull kafir 19.9 Red kaflr 19.9 Shallu 11.7 Brown kaoUane The sorghum midge did not appear this season until somewhat later than in previous years, so with the earlier maturing varieties, 1 The variety te.st hero reported is additional to an extensive series of experiments with cotton conducted by the Office of Acclimatization and Adaptation of Crop Plants and Cotton-Breeding: Investigations, the results of which will be reported elsewhere. -' T'. S. Department of AKi'iculture, Bureau of Plant Industry, Bulletin 237, entitled " Grain-sorghum production in the San Antonio region of Texas." [Cir. 120] WOKK OF THE SAN ANTONIO EXPERIMENT FARM IN 1912. 19 such as milo, plantings made as late as April 1 gave heavier yields than earlier plantings. This was due largely to the fact that there ■was a poor stand in the earlier plantings. GRAIN-SORGHUM YIELDS COMPARED WITH CORN. Two varieties of grain sorghum, Dwarf milo and Sudan durra, were grown in comparison with 42 of the most common types of Texas corn. Sudan durra gave a yield of 57 bushels per acre and Dwarf milo 63 bushels per acre, while the best strain of corn gave a yield of 40 bushels per acre, with an average of about 30 bushels for the field. Fig. 7. — Dwarf broom corn on thi> San Antonio Experiment Farm. An average yield of 657 pounds of brush per acre was obtained from plantings made on three different dates in 1912. (Photographed June 15, 1912.) DISTRIBUTION OF SORGHUM SEED. More than 150 requests for grain-sorghum seed were received from near-by farmers, and 209 packages Avere sent out to 103 of these. The varieties distributed were Dwarf milo, Sudan durra, and Dwarf Blackhull kafir. Much interest is being shown by the farmers in western Texas in the production of this grain crop, since the corn yields have been low during the past four years. BROOM CORN. In connection Avith grain-sorghum tests, three varieties of broom corn were planted, Dwarf broom corn (fig. 7) being planted on three [Cir. 120] 20 CIRCULAR NO. 120^ BUREAU OF PLANT INDUSTRY. different (lates and the others on two. The average yields from these plantings "vvere as follows : Dwarf broom corn, G. I. No. 442 057 ponntls per acre. Broom corn. (i. I. No. 243-5-^ 640 i)onnds per acre. Standard broom corn, G. I. No. 446 730 pounds per acre. The brush was of fair quality. PUBLICATIONS. As rapidly as results of a conclusive nature are secured from the experiments at the farm they are prepared for publication. In this way the facts brought out in the experimental work are promptly made available to the farmers of the region. During the year 1912 three ^publications dealing in detail with some of the j^roblems under investigation were prepared. The first of these ^ treats of the pro- duction of grain sorghum, the most important grain crop of the re- gion ; the second ^ reports the results of five years' experimentation with forage crops, and the third ^ deals with the effects produced on crop yields and soil moisture by the practice of subsoiling. Similar papers dealing with other problems will be issued from time to time as the experimental results warrant publication. 1 U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 237, " Grain- sorghum production in the San Antonio region of Texas," hy Carleton R. Ball and Stephen H. Hastings. 2 U. S. Department of Agriculture, Bureau of Plant Industry, Circular 106, " Report of the forage-crop work at the San Antonio Experiment I'arm," by S. II. Hastings. ^ TJ. S. Department of Agriculture, Bureau of Plant Industry, Circular 114, article enti- tled " Experiments in subsoiling at San Antonio," by S. H. Hastings and C. R. Letteer. [Cir. 120] [Cir. 120— C] INHERITANCE OF WAXY ENDOSPERM IN HYBRIDS WITH SWEET CORN.^ By G. N. Collins, Botanist, and J. H. Kempton, Assistant, Office of Acclimatisa- tion and Adaptation of Crop Plants and Cotton-Breeding Investigations. INTRODUCTION. As soon as European settlers in America became familiar with the different varieties of maize, the obvious classification of varieties based on the character of the seeds came into use. In very early literature we find such terms as flint, dent, and sweet referred to as representing classes already well known. The popular classification was crystallized into formal descriptions by Salisbury in 1849, but it was not until 1884, when Sturtevant extended the classification and applied as technical names Latin equivalents of the popular names, that the classification was generally recognized by scientific writers. With the exception of pod corn, which is known only as a curiosity, these groups are all distinguished by the texture or composition of the endosperm. The discovery in China of a type of maize with a new form of endosperm added another member to this long-estab- lished series.^ BEHAVIOR OF THE HYBRIDS. The waxy endosperm, characteristic of this Chinese variety, has never been observed in any American variety, while in all of the im- portations of this Chinese variety nothing but waxy seeds have been observed. When crossed with varieties that have a horny endosperm the waxy endosperm is completely recessive, the immediate result of the cross being seeds that are indistinguishable from those of horny varieties. In the second generation of crosses segregation appears to be complete, and the proportion of waxy to horny seed approximates the Mendelian monohybrid ratio, 1 : 3, with small, though significant, deviations. Detailed results of a series of crosses between waxy and horny are given in another place.^ In a general way waxy endosperm may be said to behave like sweet endosperm, which has also been found to be recessive to horny 1 Issued Apr. 5, 1913. 2 Collins, G. N. A new t.vpe of Indian corn from China. U. S. Department of Agricul- ture, Bureau of Plant Industry, Bulletin 161, 1909. 3 Collins, G. N., and Kempton, J. H. Inheritance of waxy endosperm in hybrids of Chinese maize. IV<" Conference Internationale de G^netique, Paris, 1911, p. 347-357, 1913. [Cir. 120] 21 22 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. in the first or .\eiiia generation and to segregate as a monohybrid in the following generation. Since both sweet and waxy endosperms are recessive to horny, it became of intei'est to know what n hybrid between sweet aiul waxy wonld produce. Six such crosses were made in the season of 1011. In cvei'v instance tlie I'esultine: ears were all horny, the endosj^ci'm in every way reseml)ling the horny endosperm of ordinary varieties. This synthetic production of horny endos])erm from nonhorny varieties at once suggests the idea that both sweet and waxy endo- Oa Ox s/\ sx 5X a X BX SX 5 SX mm JC sx /3 5X sx Sx 5 XLT 'X /o X Sx mm 1^ Sx sx SX aX 5; aX Hom II X sX 15 .X sx SX sx mm s sx ta. X sx mxY /<5 SX SX p 4H0m ELHOm a 5wdT ^ Hom I Hom I dWEET I mxY TOTAL '9 HOffNY 3 3tVE£T / r Fk;. 1. — DiaKiMiii '^Imwint;- the cfiniotip romposition of scconil ^'pnerntion liylirids lietwepii waxy and swict varidii's of iiiaizr. sperms represent an imperfect development through the loss or fail- ure of some element oi- hereditarv factor and that what is lackine in the sweet is supplied l)y the waxy, and vice versa. Tt has often been suggested that sweet endosperm resulted from a failure of com- plete development of horny endosperm, but no suggestion has been made regarding the nature of the deficiency. Tt now appears tliat what is lacking in a sweet xariety is supplied by the waxy, and, con- [Clr. 120] WAXY ENDOSPERM IN HYBRIDS WITH SWEET CORN. 23 verseh', that what is Licking in the waxy is supplied by a sweet variety. ^Vllile the horny seeds resulting from the cross betw^een sweet and Avaxy were indistinguishable from ordinary horny seeds, it Avas, of course, to be expected that dilferences would appear in the progeny. . Four of the six ears obtained as a result of the crosses made in 1911 were selected for planting in 1912. The ancestry of these ears is as follows: Dh22]. White Cblnese (waxy) XYoorhees Red (sweet). Dh209. White Chinese (waxy)X Bl:ick Mexican (sweet). Dh21G. White Chinese (waxy) X Black Mexican (sweet). Dh207. White Chinese (waxy)X Black Mexican (sweet). The same individual plant was not used as the parent of more than one cross. From these 4 ears 55 ears were produced in 1912. Every ear bore seeds of all three classes — horny, sweet, and w^axy. The entire 55 ears had 22,132 seeds, of wdiich 57.4 per cent were horny, 24.8 per cent sw^eet, and 17.8 per cent waxy. These proportions approximate the 9:4:3 Mendelian ratio involving two factors. Since in the first generation sweet combined wath waxy produced horny, it has been assumed that something necessary to produce horny is lacking in both the sAveet and the waxy, and an attempt has been made to analyze the behavior of these hybrids Avith this idea in vieAV. The residual factor Avhich Avhen it occurs alone is assumed to cause the sweet character Avill be called S, Avhile X Avill be used to indicate the factor of the Avaxy character. Small letters, s and £», are used to denote the absence of these factors. Since in both SAveet and Avaxy the alternative factor necessary to produce horny is assumed to be lacking, the gametes produced by sweet varieties will be repre- sented by /Sx and the gametes produced by varieties Avith Avaxy endosperm by sX. The synthetic horny Avill then be represented by a combination of these, or ^SxsX. Assuming a chance recombination of these factors in the gametes deriA'ed from these synthetic horny seeds, the gametes will be of four kinds. Both the SAveet and the waxy may be present (SX), or the sweet may be present Avithout the Avaxy (aS'j'), or the waxy without the sAveet (sX), or both may be absent (sx). At fertilization each of these kinds of gametes may unite Avith any one of the four corresponding kinds derived from the other parent, producing 10 zygotic combinations. The formation of these combinations is represented in the conventional diagram, figure 1. The four classes of gametes from one parent are given in the horizontal roAv at the top, and the same four classes from the other parent in the vertical row at the left. Each gametic combina- tion from the top is repeated four times in the squares beloAV, while [Cir. 1201 24 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. each combination at the side occurs four times in the corresponding horizontal row of squares. Thus each of the squares represents the result obtained by combining the gametes representing the horizontal and vertical rows that intersect at that point. In all cases where both S and A' occur together the seed should be horny, where only S occurs the seed should be sweet, when only A' occurs it should be waxy, and in one square (No. 16), where neither S nor A" occurs, no predic- tion can be made, since this is presumably a new condition. |^ Leaving the nature of the seed from square 16 out of consideratioj& for the ^esent, it can be seen that there are 9 squares in which both /S and A' occur (horny), 3 in which A' alone occurs (waxy), and 3 in which /S alone occurs (sweet). With respect to the horny and waxy seeds, these numbers approximate the ratios that were actually ob- tained, there being roughly 9 horny seeds to 3 waxy, but the sweet seeds occur as 4 instead of 3 out of every 16. We must therefore assume that the new type, sxxx, represented in square 16 and contain- ing neither the sweet nor the waxy factor is or resembles sweet. Careful scrutiny of the sweet seeds failed to show any consistent differences that would allow another class to be separated, but if the jDresent method of looking at the cross is to be of use it should be possible to detect the differences between the ordinary sweet seeds and this new class by analytical breeding. One way to test the theory that this new class of apparently sweet seeds really lacks the element ordinarily concerned in the production of the sweet character is to cross it with pure waxy. The cross of sweet and waxy has heretofore always produced horny, but if this new class lacks the factor for sweet, the cross should have the gametic composition represented by squares 12 and 15 and should result in waxy instead of horny seeds. Another test could be applied by cross- ing the new class with ordinary horny varieties. The first genera- tion of this cross should be horny, but the formula of gametic com- position, /S'A'5',7!, would be the same as for the synthetic horny, and in the second generation 3 out of every 16 seeds should be waxy. Thus, if the present assumption regarding the nature of this class is correct we may expect two apparent anomalies, not observed hith- erto, viz, waxy seeds as the immediate result of crossing sweet and waxy, and waxy seeds in the second generation of a cross between sweet and horny. COMPARATIVE WEIGHT OF HORNY, SWEET, AND WAXY SEEDS. The view that waxy as well as sweet endosperm may be compared to incomplete stages in the production of the horny endosperm is strengthened by the relative weights of the sweet, waxy, and horny seeds where all three classes occur on the same ear. The horny seeds [Cir. 120] WAXY ENDOSPERM IN HYBRIDS WITH SWEET CORN, 25 are the heaviest, followed by the waxy and sweet. A total of 3,161 horny seeds weighed 790.3 grams, 2,387 waxy seeds weighed 565 grams, and 5,247 sweet seeds weighed. 1,028.3 grams. Eedifced to W' eight per seed, these results give 0.253 gram for the horny^ 0.237 gram for the waxy, and 0.196 gram for the sweet. The average weight of the four classes, white horny, white waxy, w^hite sweet, and colored sweet, was determined on 49 ears belonging to four families.^ The results are shown in Table I. Since the average weight of seed differs on different ears, the comparative weights of the different classes are expressed in terms of the weight of the horny seeds. Table I. — Weight of irliite warij, colored sweet, and irhite sweet seeds of maize. Family. Number of ears. White horny. White waxy. Colored sweet. White sweet. Dh221 12 13 15 9 100 100 100 100 93. 8± 1.02 9.5. 6± .77 96. 5± .95 91. 2± .93 77. 7± 1.52 76. 2± 1.71 79. 3± .94 76. 8± 1.20 77. 4± 1.40 Dh209 76. 4± 1.45 Dh216 77. 8 ±1.28 Dh207 72. 4± 1.90 Total 49 100 9.5. 2± .45 77. 6± .65 76. 4± .71 There appears to be no real difference in weight betAveen the white and colored sweet seeds. The sweet seeds are, however, consistently lighter than the wax}^, which in turn are lighter than the horny. The average difference betw^een the sweet and waxy seeds is 22.5 per cent of the weight of the sweet; that between the waxy and horny is 5 per cent of the waxy. The volume and specific gravity of the classes of seeds were also determined in eight of the ears. The results showed that the differences in w^eight were associated with corresponding differences in volume, the specific gravity being the same for all classes. COMPARISON OF RESULTS WITH MENDELIAN EXPECTATION. In the previous pages the classes occurring in the second generation of the sweet X waxy cross have been referred to as approximating a 9:4:3 ratio. While this approximation is so close as to exclude the possibility of finding any other ratio that will more closely fit the ob- served number, the deviations are too large to be ascribed to chance. Table II gives the number of horny, sweet, and waxy seeds com- pared with the number expected in accordance with the above ratio. For both the horny and waxy it will be seen that the deviation for the entire series is more than five times the probable error. In the case of the sweet seeds the total deviation is no greater than should 1 The two classes, colored horny and colored waxy, could not be weighed accurately, since these seeds had been clipped in order to determine whether they were horny or waxy. [Cir. 120] 26 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. be e.xpoctcd. hut ;m examiiiatioii of the separate families indicates that this aijreeiiient is accidental, for of the four families one is above the expected innnber l)y nearly five times the |)robable error and two are below by more than three times the probable error. Tablk II. — Proportiiin of xirccf. ini.iii. niiil lioniji ncctls of iiKiiic coDiiinrrd irith the (•■iiKctcd ratios. Descriptive data. Family. Total. ])h22I. Dh209; Dh216. Dh207. Number of ear.s 12 6,033 1,399 1,.'-,0S±22. 7 -109 4.8 23.2 1,085 1,131 ±20. 4 -46 2.3 18.0 3,519 3, 394 ±25. 9 +155 6.0 58.8 14 5,769 1,3.58 1,442 ±22. 2 -S4 .3.8 23.5 1,059 1,082±20.0 -23 1.1 IS. 4 3,352 3,245±2.5.3 + 107 4.2 .58.1 17 6,413 1,720 1,603+25.4 + 117 4.6 26.8 1,132 1,2(J2±21.1 -70 3.3 17.7 3,. 561 3,608±26.8 -47 1.8 55.6 12 3,917 986 979 ±18. 3 +7 .04 25.1 665 734 + 11.5 -69 6.0 17.0 2,266 2, 204 ±20. 9 + 62 3.0 .57.8 55 Number of seeds 22, 132 Sweet seeds: Number 5,463 Expected number 5, ,5:53+43.4 Deviation -70 Deviation divided by proba- ble error L6 Per cent 24.7 Waxy seeds: Number 3,941 Expected number 4,149+39.1 Deviation -208 Deviation divided by proba- ble error 5.3 Per cent 17.8 Horny seeds: Number 12,728 Expected number . . . 12,450+49.6 Deviation +278 Deviation divided by proba- ble error " Per cent 5.6 .57.5 CONCLUSIONS. The general classification of maize varieties is based on the char- acters of the seed. Varieties with horny, soft, and sweet endosperm have been recognized since early colonial times, and these three types represent the only forms in which the endosperm of maize was known to exist until the di.scovery of w^axy endosperm in a variety of maize introduced from China in 1908. Investigations of the heredity of this new type of endosperm show^ that Avhen the Chinese variety is crossed with sweet varieties seeds are produced which, unlike either parent, have a horny endo.sperm. Although this synthetic horny endosperm is indistinguishable from that of ordinary field varieties, it behaves in an entirely different manner in the following generation. All of the plants grown from these hybrid seeds produced ears with horny, sweet, and waxy .seeds. The jn'oportion in which the three classes occurred approximated nine horny, four sweet, and three waxy, a ratio indicating that the sweet and waxy characters are the result of independent factors and that one out of every four sweet seeds, though indi.stingui.shal)le from oi'dinary sweet seeds in the first generation, represents a new type of sweet corn. Experi- ICir. 120 J WAXY ENDOSPERM IN HYBRIDS WITH SWEET CORN. 27 inents have been oulliiiecl fur isulatiii<^- and determinin«>- the nature of this new type. Since all three classes of seeds occurred on the same self-])ollinated ear, it was possible to compare the weights of the three classes where the parentage of all was identical. The horny, waxy, and sweet seeds from each of 40 ears were separately weighed. The result shoAved that in nearly every ear the horny seeds w^ere heavier than the waxy, which were in turn heavier than the sweet. Specific- graA'ity determinations of the different classes showed no significant differences. The differences in weight must therefore be due to dif- ferences in size. [Cir. li:u] [Cir. 120 — D.] LEAF-CUT, OR TOMOSIS, A DISORDER OF COTTON SEEDLINGS.^ By O. F. Cook, Bionomist in Charge of Croi) Acclimatization and Adaptation Investigations. INTRODUCTION. Seedlings and young cotton plants are subject to a peculiar disorder .that results in extensive injury to the leaves and frequent abortion of the tenninal bud. While these injuries are seldom fatal, they undoubtedly impede the growth of the plants, delay the period of production, and reduce the crop. Though not taken into account hitherto, the losses occasioned by the leaf-cut disorder, though most severe in the Southwestern States, seem to be very general and must amount to millions of dollars every year. With the advance of the boll weevil this form of injury to the young plants acquires a more serious aspect, because it is of the utmost importance to shorten the period of production in order to avoid damage by the weevil. The leaf-cut handicap can be reduced by improved cultural methods, as stated in previous publications of the Bureau of Plant Industry.- LEAF-CUT DISTINGUISHED FROM LEAF-CURL. Leaf-cut is suggested as a popular name because mutilations of the leaves are the most characteristic symptom of the disorder. The name '"' juvenile leaf-curl " has been applied in previous publications, but is inconveniently long and not sufficiently distinctive. Moreover, the leaves of cotton seedlings are subject to another malformation, induced by plant lice, for which the name " leaf -curl " is more appro- priate. Leaf-cut is very widely distributed and familiar to planters, though generally confused with the leaf-curl caused by plant lice. Though both forms of injury are likely to be found in the same field, or even on the same individual plant, they are easily distinguished. (Fig. 1.) The leaf-curl is a crumpling or arching up of the leaf between the veins, but without perforations or rents. Even in cases of great dis- tortion by leaf-curl the tissues of the leaf are left entire, without 1 Issued Apr. 5, 1913. 2 See Report of the Chief of the Bureau of Plant Industry, U. S. Department of Agri- culture, for 1011, p. 24 ; for 1912, p. 27 ; and U. S. Department of Agriculture, Bureau of Plant Industry, Circular 96, p. 13. [Cir. 120] 2^ 30 CIRCULAR NO, 120, BUREAU OP PLANT INDUSTRY. being iDunctured or torn. Leaf-cut injuries, on the other hand, rep- resent an actual destruction of some of the tissues of the leaf, leav- ing irregular holes or marginal incisions. In single words, leaf-curl may be described as distortion of the leaves, leaf -cut as mutilation. In allusion to this distinction the word " tomosis " is proposed as a technical name for the leaf-cut disorder, while the distortion caused Fig. 1. — Young plant of Egyptian cotton yiown at Lanham. Md. Tho lowcf li>aves arc affected l).v leaf-cut and upper loaves by Ipaf-curl. (Natural si/c. i by the plant lice may be called '' hybosis." The insects inhabiting the badly distorted upper leaves of figure 1 were identified by Mr. Th. Pergande, of the Bureau of Entomology, as Aphis gossyjnl Glover. And in addition to the open wounds that result from leaf-cut there are usually some that have healed, giving a characteristic torn-and- mended appearance. Such scars, lil\e other leaf-cut wounds, often [Cir. 120] LEAF-CUT, A DISOEDER OF COTTON SEEDLINGS. 31 lie in a somewhat radiating; position between the principal veins. Healing of wounds and regeneration of lost parts show that the injuries are liable to occur at a very early stage in the development of the leaf. Sometimes an extensive new growth or regeneration takes place, resulting in a curious doubling or overlapping of lobes of injured leaA'es. The power of the injured tissues to heal is also responsible for adhesions between parts that lie folded together in the bud. These adhesions are usually responsible for failure of normal expansion of the blade. None of these secondary symptoms occur Avith the leaf-curl induced by plant lice. CAUSES OF LEAF-CUT INJURIES. Leaf-cut is hardly to be reckoned as a disease unless the word is used in its most general application that includes any departure from normal structure or function. Neither of the two general classes into which diseases are usually divided, constitutional and parasitic, would include the leaf-cut. Though some of the cells are destroyed, the remaining tissues of the plant do not become abnormal in any way, and there is no indication that parasitic organisms of any kind — bacteria, fungi, insects, mites, or wonns — are involved. Another class of ecological disorders may need to be recognized, intemiediate between physiological diseases and mechanical injuries or trauma- tisms. Leaf -cut is a disease only in the sense that frostbite, snow blindness, and other environmental injuries are to be considered as diseases. Young cotton plants are often subjected to extreme conditions during the early stages of gi'owth, when the leaf-cut injuries occur. The leaves and roots are still close to the surface soil, where they can be chilled at night and scorched in the daytime. Cold nights are sometimes looked upon as the cause of the injury, and may be an intensifying factor, but the sudden heat of a bright morning sun seems more likely to kill the cells of the young leaves than low tem- peratures during the night. Leaf-cut often affects late plantings long after the night temperatures have ceased to approach the freez- ing point. It has been noticed that exposure to a bright morning sun after a cold night will tlirow cotton seedlings temporarily into a wilted state, doubtless because the leaves lose Avater by transpira- tion faster than it can be absorbed by the chilled roots. Leaf-cut seems to be especially prevalent under such conditions. That leaf-cut is in some way connected with exposure or wilting of the delicate tissues is also shown by the fact that tlie injuries are most severe and occur most frequently along radiating lines midway between the principal veins. These lines of greater susceptibility represent the most exposed parts of the upper surface of the young [Cir. 120] 32 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. leaf as it lies folded in the bud. The only suggestion foi- explaining the very irregular manner in which the cells are killed is that some of them may be unable to complete their divisions and nuclear read- justments during the night and may thus be left in an unusually susceptible condition. Sections of injured leaves prepared by. Dr. Albert Mann, of the Bureau of Plant Industry, show that nuclear and protoplasmic disintegration are the earliest symptoms. The damage often begins with the death of a single cell, which results, of course, in increased exposure for the neighboring cells. Plants protected by partial shade suffer less than those that are fully exposed, but, on the other hand, full exposure does not induce leaf-cut when the plants are growing on wet land, where the surface remains moist and is kept cool by evaporation. The moist atmos- phere and partial shade afforded by ordinary greenhouse conditions also afford complete protection from leaf-cut. Even in parts of the same field there may be obvious differences in the extent of leaf-cut injury. Plants that stand close together often show much less injury than more scattering plants in the same rows. AAliere the soil is too dry to germinate all the seed the leaf -cut injuries are more extensive. Such differences indicate the possibility of avoiding or reducing the damage from leaf-cut by giving better attention to the seed bed and to methods and times of planting and thinning. ABORTION OF TERMINAL AND AXILLARY BUDS. Though mutilation of the leaves is the most frequent and familiar symptom of the leaf-cut disorder, abortion of terminal buds is a more serious injur}^ In severe cases of leaf-cut, from 30 to 60 per cent of the plants have been found with their terminal buds aborted. When the leaf-cut injuries are confined to the individual leaves the effect is merely to retard the growth of the plant, but when the termi- nal bud is lost the plants are permanently deformed and usually produce a much smaller and later crop than normal individuals in the same field. In the most severe form of the disorder the 3"oung seedlings lose the buds in the axils of the cotyledons as well as the terminal bud. Such plants are unable to form any true leaves, but the cotj'ledons increase in size and the hypocotyl becomes much thickened. In some cases the root begins to form a subterranean shoot, like those that develop vegetative buds when plants have been killed to the ground in the winter. When abortion of the bud takes place higher up, so that the plants have one or two true leaves, the blades grow much larger than usual and the petioles become greatly elongated. If thinning be deferred until the normal planta are 10 inches or a foot [Cir. ll'Oj LEAF-CUT, A DISORDER OF COTTON SEEDLINGS. 33 high it is easy to distingiiiKli and remove the deformed individuals and leave only the healthy and vigorous ones. Under the usual plan of thinning the cotton early it is much more difficult to recognize and remove the injured plants. SUSCEPTIBILITY AND IMMUNITY. Susceptibility to leaf-cut is usually limited to the seedlings and young plants less than 10 inches high. Sometimes the change from susceptibility to immunity is Aery abrupt. Plants that have had every leaf injured up to the sixth or eighth may then begin to put out entirely uninjured leaves. These abnipt changes may affect Avhole roAvs or fields of cotton, as if the later uninjured vegetation had groAvn out after a hailstorm. Whether the plants become im- mune to leaf-cut simply because larger stature carries the new groAvth farther away from the oA^erheated soil, or because a deeper root system affords a more regular supply of moisture, or because the Aveather conditions become more uniform as the season advances has not been determined. All these factors may cooperate, or there may be others as yet unsuspected. A few cases of abnormal individuals have been observed where injuries Aery similar to leaf-cut continued during the Avhole life of the plant. Some of these plants Avere hybrids and others Avere mu- tations, but all of them Avere abnormal in other Avays, as Avell as in the irregular texture of the foliage. It seems not unreasonable to suppose that abnormal plants should remain more susceptible to any external conditions that have adverse effects upon the activities of the cells. Though all the different types and varieties of cotton seem to be susceptible to leaf-cut injuries, certain differences have been noticed. The leaA^es of the Durango cotton and other Upland varieties are often injured much more seriously than those of Egyptian cotton in adja- cent roAvs, but at the same time the Egy^ptian cotton may show a larger percentage of abortion of terminal and axillary buds. The immunity may lie in the improA^ement of conditions rather than in an increased resistance on the part of the plant. With the plant-lice injuries there is a gradual reduction of the amount of distortion that the insects are able to produce, Avhich may indicate the deA^elopment of a different kind of immunity in this disorder. It is true that the plant lice usually disappear as the season advances, but even AAdien the insects remain abundant the distortion becomes less as the plants groAv larger. CONCLUSIONS. Leaf-cut is a disorder of cotton seedlings characterized by mutila- tion of the leaves and abortion of the terminal buds. Leaf-cut has 34 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. been confused Avith the distortion of the leaves by phuit lice, but the two malformations are readily distinguished. Leaf-cut is in the nature of an environmental injury, not due to l^arasitic organisms or to constitutional weakness, but api:)arently connected with exposure to heat and dryness. All varieties of cotton are susceptible during the early stages of growth. Though leaf-cut is not fatal, it is responsible for much damage by retarding the growth of the young plants. The loss of the terminal buds interferes with normal habits of branching, and the plants are permanently deformed. Damage from leaf-cut can be avoided or reduced by improved cultural methods, and the deformed plants can be removed by later thinning. [Cir. 120] ADDITIONAL COPIES of this publication -t\. may be procured from the Superintend- ent OF Documents, Government Printing OflBce, Washington, D. C, at 5 cents per copy U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— Circular No. 121. WILLIAM A. TAYLOR, Chief of Bureau. MISCELLANEOUS PAPERS. The Culture of Durango Cotton in the Imperial Valley The Control of the Sugar-Beet Leaf-Spot The Work of the Huntley Experiment Farm in 1912 Methods of Securing Self-Pollination in Cotton ARGYLE McLACHLAN V.W. POOL and M. B. McKAY DAN HANSEN R. M. MEADE Issued April 12, 1913. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1913, [Cir. 121] 2 BUREAU OF PLANT INDUSTRY. Chief of Bureau, 'William A. Taylor. Assistant Chief of Bureau, L. C. Corbett. Editor, J. E. Rockwell. Chief Clerk, James E. Jones. THE CULTURE OF DURANGO COTTON IN THE IMPERIAL VALLEY/ By Argyle McLachlan, Assistant in Crop Acclimatization, Office of Western Irrigation Agriculture. INTRODUCTION. Diiranojo cotton is a long-staple Upland variety recently intro- duced by the United States Department of Agriculture and appar- ently well adapted to Imperial Valley conditions. A general account of the variety and of its behavior in the Imperial Valley in compari- son with other long-staple sorts has been given in a recent publica- tion. ^ The object of the present paper is to call attention to spe- cial methods and precautions that need to be observed in the devel- opment of a long-staple industry under the local conditions. Most of the planters have grown only short-staple cotton in the past and many are not familiar with cotton culture or with methods of irrigation. Durango cotton was first grown in the Imperial VaUey in 1911 by Mr. W. E. Wilsie on 3 acres of land near El Centro. In 1912 about 200 acres were planted by six growers from seed grown by Mr. Wilsie. Following the favorable results obtained in 1911 and 1912, it appears that between 6,000 and 7,000 acres will be planted to Durango cotton in the Imperial Valley in 1913. All the available supplies of Durango seed in Texas were purchased. A much larger acreage would be planted if more seed could be obtained. It is also estimated that about 20,000 acres more will be planted to short-staple cotton, chiefly of the Mebane (or Triumph) variety. This total estimate of 26,000 or more acres includes acreages to be planted in Lower Cali- fornia adjoining the Imperial Valley, which are operated chiefly by residents of the valley and from which the crops are marketed through Imperial Valley points. In undertaking the growing of this greatly increased acreage of Durango long-staple cotton, it should be appreciated at once that better care of seed and more care in cultivation, harvesting, ginning, and handling are required in producing cotton with long and uni- form lint than merely for producing short-staple cottons. > Issued Apr. 12, 1913. 2 Cook, O. F. Durango cotton in the Imperial Valley. In U. S. Department of AgriculUm', liuieau of Plant Industry, Circular 111, 1913. [Cir. 121] 3 4 CIKCULAR NO. 121, BUREAU OF PLANT INDUSTRY. The most essential step in connection with a long-staple cotton industry is the establishment and maintenance of clean and select seed of the variety. Deterioration of a variety from failure to select the seed or through carelessness in keeping it free from mixture with seed of other varieties can not be overcome by cultural or other methods. The cotton industry of the Imperial Valley will be best served if proper means are taken during the season of 1913 to insure a supply of select, pure Durango seed for planting in 1914. CLEAN SEED FOR PLANTING IN 1914. MIXED SEED SHOULD BE DISCARDED. The great promise of the Durango variety has led the cotton growers to plant this year all available seed, though it is known that a large proportion of it is mixed or impure and can be used only for a single season. At least three-fourths (or about 4,500 acres) of the Durango cotton acreage in the Imperial Valley m 1913 will be planted to seed which was grown in the valley in 1912. Unfortimately, this seed has been mixed and cross-pollinated with short-staple cotton. The seed planted in 1912 was grown in the valley in 1911 on clean land which had been in alfalfa the preceding season, so that no mixture occurred in the field. A small amount of mixture (possibly 2 to 4 per cent) occurred in the ginning of the Durango seed in 1911. The mixing probably took place in the screw conveyor which carries the seed from the gin stands to the wagon. The most serious mixture occurred in the fields in 1912. A large part of the Durango cotton was planted on land which had been in short-staple cotton in 1911, and the volunteering of the previous season's variety caused an actual mixture of short-staple plants in the Durango fields amounting to cS to 20 per cent. This has resulted not only in an actual seed mixture in the crop produced in 1912, but also, and what is even more serious, in the hybridization of the Durango with short-staple cotton. Either sort of mixture, if allowed to persist, would prove disastrous to the new long-staple industry. Seed mixture alone in stocks of valley seed might be successfully overcome, but the cross-pollination of the Durango by short-staple varieties is a mixture practically impossible to eradicate. While some of the valley Durango seed to be planted in 1913 contains not more than 5 per cent of short-staple seed, the hybridization which has taken place makes it essential for the eventual good of cotton growing in the Imperial Valley that all mixed seed be sent to the oil mill or some other seed-destroying agency this fall, so that no stock of mixed Durango seed can be planted in 1914. ICir. 121 J CULTURE OF DURANGO COTTON IN THE IMPERIAL VALLEY. 5 Fortunately, a small quantity of uncontaniinated Durango seed (enough to plant about 700 acres) has been secured from Texas for 1913 plantmg, and a small quantity was grown in the Imperial A'alley mider conditions that warrant its purity. If handled properly in 1913, the crop from tliis seed will yield an abundance of clean seed for all local needs in 1914. RESPONSIBILITY OF PLANTERS OF CLEAN SEED. Those who plant clean seed m 1913 should assume the responsi- bility of providmg clean seed for planting in 1914. The practical bearing of this responsibility of the mdividual planter is in the fact that the establishment of a long-staple mdustry depends largely on the production of large quantities of first-class Durango cotton. The production of 10,000 or 20,000 bales of such cotton in 1914 and succeeding years would mean much to the Imperial Valley. The man who, through carelessness in 1913, allows clean seed planted on 50 or 100 acres to become mixed may cut down the production of first-class Durango cotton in 1914 to the extent of 2,500 to 5,000 bales. It is therefore important that each grower take it upon hun- self to keep his clean Durango seed from becoming mixed with other varieties. PRESERVATION OF CLEAN SEED. CLEAN LAND. In the first place, all clean seed should be planted on clean land. If planted on land cropped to cotton hi 1912 there is every proba- bility that mixture will be occasioned by the volunteering of seed scattered from the previous crop. Up to the present there has arisen no necessity for the eradication of volunteer cotton, so that methods to this end have not yet been devised. The condition of cleanness of fields on which to plant pure Durango seed should be absolute m order to msure the deshed end of raismg clean Durango seed. A few volunteer plants of another variety in the field, along borders, or in fence corners are as dangerous as a great number. ISOLATION. Further, plantmgs of clean Durango seed should be made in fields isolated from plantings of other varieties of cotton, in order to guard agamst cross-pollmation of the Durango by insects. If it proves impracticable to isolate the field of clean Durango by dis- tance (of one-eighth mile or more), the precaution should be taken to plant a wide strip (50 feet or more) of milo or of some other tall- growmg crop between the Durango and other cottons. [Cir. 121] 6 CIRCULAR NO. 121, BUREAU OF PLANT INDUSTRY. MINOR DETAILS. It would seem unnecessary to call attention to the danger of mix- ing Durango seed with other varieties at planting time, but acquaint- ance with carelessness in the small details of cotton handling makes it apparent that warnmg shoukl be sounded against the possibDity of mixing in each operation connected with the planting and growing of the crop. This is important in cases where the Durango and Mebane cotton are planted on the same farm. Sacks containing Durango seed should ho marked "Durango." The handling of the seed should not be hitrusted to incompetent hired help, who might not appreciate the importance of such details. The planting machine should be thorouglily cleaned of any seed which may have become lodged while planting seed of another variety. PROPER TILTH. Instances have occurred in the planting of cotton in the Imperial Valley where replanting has been necessary because of poor tilth. It would be a serious loss to the industry if any of the clean seed were to fail to yield a good stand through lack of thorough and special preparation of the land. There is a very limited supply of clean Durango seed and none with which to replant. CLEAN CONVEYANCES. Special care to prevent mixing must be exercised at picking time as long as other varieties of cotton are being harvested in proximity to the Durango or even in the same valley. Wagon racks in which other cottons have been hauled should be cleaned of stray bits of seed cotton before loading Durango cotton. Cars in which Durango seed cotton may be transported to the gins should likewise be properly cleaned. The identity (number and name) of the car in which the clean Durango seed cotton is carried should be clear to grower and ginner alike immediately on shipment. The producer should accom- pany his clean Durango seed crop to the gin and assist the ginner in seeing that the conveyors to the gin stands, the feeders, and the cleaners are thoroughly cleaned of other cotton seed previous to ginning the Durango. The aprons of the gin stands should be re- moved and the seed sacked from the floor in front of the stands in- stead of being caught in wagons after passing through the screw con- veyor beneath the stands. One can not honestly claim that his seed is salable for planting purposes unless he takes every precaution to preserve its purity. HANDLING PLANTINGS OF MIXED SEED. A certain responsibility to the cotton industry of the Imperial Valley rests on those who plant mixed Din-ango seed. Their imme- diale responsibility consists in doing their utmost to ])rovide first- (Cir. 121] CULTURE OF DURANGO COTTON IN THE IMPERIAL VALLEY. 7 class Durango fiber from their mixed crop in order to increase the outturn of high quality cotton from the valley. This can best be accomplished by removing the short-staple plants from the mixed Durango field previous to picking time. The eradication of the Mebane plants and plants of other varieties of cotton from plantings of mixed Durango is entirely feasible, as Durango plants display characteristics of leaf and other features wPdch distinguish them throughout theh growth fi'om plants of other varieties. Durango leaves have narrower and more sharply pointed lobes than Mebane leaves, and appear more definitely five lobed. Growers should take the first opportunity presented to compare Durango and Mebane cotton plants in order to become acquainted with these differences. It will be possible to distinguish and cut out many of the plants of other varieties at chopping time. By persistent effort all short-staple plants can be eradicated from the mixed Durango fields. This will leave in the fields only Durango and some hybrid plants which wall yield much higher quality fiber than fields from which the short-staple plants have not been removed. There is no reason why fields of mixed Durango, if handled in this manner, should not yield fiber this season of as good character as that from fields of clean Durango. PAST EXPERIENCE GUARANTEES PROGRESS. A distinct advance has been made by the cotton-growing com- munity of the Imperial A^ alley since the planting of "Georgia" cotton in 1909. The advance thus far accomplished amounts to the estabhshment within a period of four years of a short-staple industry based practically on the best obtainable variety of short-staple cotton. The experience alread}^ gained through the ehminating from the industry in the Imperial Valley of numerous varieties of cotton which have been tested and proved poorly adapted to the conditions should make it possible for the cotton growers to take the necessary steps in connection with a long-staple industry for main- taining clean stocks of seed of one accepted variety and of providing proper culture to produce superior fiber. PROPER CULTURE TO PRODUCE GOOD FIBER. Good Durango fiber, 1-|^ to 1^ inches long, of even length and of good strength, can be grown by providing the plants with normal growing conditions through the season. Water should be apphed in such a way as to allow the plants to grow and fruit normally with- out being subjected either to drought or overwatering or to undue checking or acceleration of growth. Either checking or accelerating growth has a direct effect in causing diversity in the length and [Cir. 121] 8 CIECULAR NO. 121, BUREAU OF PLANT INDUSTRY. strength of fiber. Irregularity of any sort is of much more serious consequence in the case of long-staple than in short-staple cottons.^ Water should be given the jjlants as they require it rather than at stated intervals. The grower should be able to tell by the a])pear- ance of the plants when they need water. Slight waiting at midday during very hot weather is not an indication that the plants are suffering for water if they freshen up in the evening and morning. It should be kept in mind by all cotton growers that overwatering may cause damage from wliich the plants mil be unable to recover because of the destruction of the tip buds of the stem and branches which continue the growth. Land containing a quantity of humus, such as lands wliich were heavily covered with brush previous to reclamation or old alfalfa sod lands, wiU produce Durango cotton of better character than lands which are poor in humus or which have been previously over- cropped. The cotton industry of the Imperial Valley would profit greatly if all Durango cotton this season could be planted on alfalfa sod land. Not only would this guarantee a larger crop and more clean seed for use in 1914, but better and more valuable fiber. It can be foreseen that not all who plant Durango cotton this season will produce a satisfactory crop. The best grown Durango, which will yield over a bale per acre of excellent fiber, must be used as a basis for judging the crop value of Durango cotton. It must not be lost sight of that the failure to plant on good land or the fail- ure to water in a way to provide normal growth is conducive in the case of long-staple cotton not only to shortage of the crop, but to unevenness and shortness of the fiber. PREPARATION OF LAND. Success has been attained by following widely different methods of planting, cultivation, and irrigation of cotton in the Imperial Valley. This is largely due to the great variations in soil and partly, no doubt, to the fact that water for irrigation has been practically unhmited as yet. It has been possible to apply sufficient water to grow a crop of cotton without conserving the moisture by cultivation. The growing of Upland cotton under irrigation is at best still in the experimental stage. It is highly probable that methods of cotton culture in the Imperial Valley may be materially changed during the next few years. Changes which will bring about the production of a uniform quality of cotton will be of great value to the industry. The following cultural suggestions may serve to promote proper adjustments in the methods of cotton culture under irrigation. ' Cook, O. F. Cotton improvement under wee\'il conditions. V. S. Department of Agriculture, Farm- ers' Bulletin ."ifll, p. 14-1.'). Cook, O. F. Results of cotton experiments in 1>)1 1. IT. S. Department of Agriculture, Bureau of Plant Industry, Circular 96, p. 6-7. fCir. 121] CULTURE OF DURANGO COTTON IN THE IMPERIAL VALLEY. 9 PLOWING. Irrigated lands should be thoroughly and deeply stirred in prepara- tion for planting cotton in order to insure necessary aeration and to put them into condition for the storage of water and the proper growth of the plants. Some "soft" lands containing an abundance of humus may not rec^uire more than a thorough disking before being planted for the first time, but after a season of settling" and packing by irrigation all characters of soil will be benefited by thorough plowing. DISKING AND HARROWING. Following heavy irrigation, soft lands can be put into condition to plant flat by disking and harrowing. This saves the labor of hsting. The seed is planted in the moist soil and germinated without irriga- tion after planting. It is much more difficult to handle heavy soils in this way. Usually following irrigation the heavy soil is hsted and the seed is planted on the ridge and germinated by mmiing water through the furrows. FURTHER PREPARATORY OPERATIONS. Land that is to be planted flat should be nmlched by harrowing, to conserve the moisture and keep the land soft until planting time. On land that has been fisted the ridges should be pulled down by cross- harrowing, to leave the tops flat and broad and mth the soil on top well pulverized. It should be kept in mind that all the operations of land preparation are for the purpose of securing fidl and prompt germination by providing the seed with proper conditions of soil and moisture. PLANTING. PLANTING TIME. As a general rule the best results will be had from planting after March 20. Earfier plantings may be subjected to numerous setbacks, amomiting in some instances to a complete loss of the stand. AU cotton should be planted by the last of April, though good crops have been grown from May plantings. DEPTH. Planting is done with a 1-row or 2-row machine planter. If only a very limited quantity of choice seed is available a larger acreage can be planted by dropping the seed by hand, four or five to the hill. The seed should be covered to a depth of 1^^ to 2 inches. DISTANCE. It is usual to plant lows from 3 feet 6 inches to 3 feet 10 inches apart. 85941°— Cir. 121—13 2 10 CIKCULAR NO. 121, BUEEAU OF PLANT INDUSTRY. LATE THINNING AND CLOSE PLANTING. Plants are usually thinned to stand 18 inches or more apart in the row when they are about 5 inches tall. Late thinning, when the plants are about 10 inches tall, to leave them 8 to 12 inches apart will in all probability provide better conditions for cultivation and harvest, as well as a larger crop.* Durango cotton adapts itself especially to this plan. From the results of tests of the new method of thinning it would appear that marked progress can be expected in cotton culture by the introduction in the Imperial Valley of late thinning to close distances. It is suggested that cotton growers in the valley try this new method on a few acres of their plantings. CULTIVATION. HARROWING YOLTNG COTTON. Crusted soil, occasioned by rains or by irrigation water covering the rows while the plants are small, should be thoroughly stirred by harrowing across the rows. This treatment is of great benefit to young plants and \vill usually save the stand if the seed is in process of germination. INTERTILLAGE. Thorough cultivation of a nature to maintain a thick mulch on the surface should be practiced following irrigation as long as the plants will permit passage between the rows. This will insure a steadier and more even supply of water to the plants and thus a more normal growth. IRRIGATION. IRRIGATION BY FLOODING. Flat irrigation, which allows the water to stand against the plants, is not necessarily harmful on soft lands, where no caking of the soil can occur about the plants, unless the water is left long enough to scald. Flat irrigation on hard land has the disadvantage of causing a caking of the soil, which is especially undesirable about young plants. INFREQUENT EARLY IRRIGATION. Irrigation should be deferred as long as possible after the cotton is up to stand. It is undesu'able to crowd the young plants by early irrigation into excessive vegetative growth, which will later have to be sustained by heavy and frequent irrigations and will result in a crop difficult to pick because of oversized plants. Other disad- vantages result unless precaution is taken to control the growth of plants when young. ' See U. S. Department of Agriculture, Bureau of Plant Industry, Circular Ho, article entitled "A new system of cotton culture," by O. F. Cook. [Clr. 121] CULTURE OF DUKANGO COTTON IN THE IMPERIAL VALLEY. 11 DETERMINING WHEN TO IRRIGATE. The grower should watch the growing tip of the cotton plants for indications that the plants need water. While a general mlting in the hottest part of the day does not necessarily mean that the plants are suffering for water, wilting of the young leaves of the growing tip or stoppage of gi'owth is usually a clear indication that the plants are getting too much or too little water. Long-staple cotton will be most successfully grown by those who apply water when the plants indicate that they need it in order to make uninterrupted growth. PLANTING IN FURROWS AND VOLUNTEERING. Plantmg in shallow furrows may find its place m cotton culture in the Imperial Valley in connection with the volunteermg of the crop. The soil gradually worked toward the plants in the process of cultivation will cover the lower buds of the stem, so that when cultivation is completed in midsummer one to three of the buds "will be completely protected. This protection should preserve the buds from killmg frosts m the whiter, so that prompt and uniform volunteering can be brought about in the spring by turning the soil away from the old stumps. Otherwise, volunteering is a slow and uncertain process. CLEAN PICKING. Cleaner picking and handlmg of long-staple cotton should be prac- ticed more than has been the custom in connection with short-staple cotton. The advantage of clean grade has been appreciated in the sales of short-staple cotton during the past season. It is even of greater comparative value m the case of long-staple cotton, as low grade occasions greater comparative loss in value in long-staple than in short-staple cotton. Cleaner picking can be insisted on. The seed cotton in the field can be dumped on big canvas wagon sheets instead of on the ground to gather more trash and grit, or mto wagons standmg in the field, and it can be covered at night to exclude blowmg sand. Any expe- dient used to provide higher grade cotton is of distinct advantage to the grower. CENTRALIZED GINNING. If a central ginning plant is provided for ginning Durango cotton exclusively, it will be very desirable that all Durango cotton in the valley be sent there. The equipment should be suited to handling the long-fibered cotton in a way to preserve its value and to turn out high grade. Less difficulty will be found in preserving clean stocks of seed, a more uniform and superior bale can be turned out, and centralization will facilitate sale. All these items are of advan- tage to the grower. [Cir. 121] 12 CIBCULAE NO. 121, BUREAU OF PLANT INDUSTRY. SUMMARY. Durango cotton is a long-staple Upland cotton recently placed in commercial culture in the Imperial Valley through the distribution of seed by the United States De])artment of Agriculture. It pro- duces an even fiber 1^ to 1^ inches long, }delds heavily, and turns out 31 to 32 per cent of lint. It has large bolls and can be picked about as easily as short-staple cotton. More than 6,000 acres will be planted to Durango long-staple cot- ton in the Imperial Valley in 1913. About 4,500 acres will be planted to mixed valley-grown seed. The remaining acreage will be planted to Durango seed brought from Texas, some of which is known to be free from mixture with other varieties of cotton. About 20,000 acres will be planted to short-staple cotton in the same valley. A long-staple cotton industry based on Durango cotton can be established only by maintaining clean and select stocks of Durango seed and hj using the proper cultural methods to provide normal gi-owth for the plants and fiber. A gi'eat advantage can be gained by better methods of picking and handling the mature crop. Each individual who plants clean Durango seed tliis spring should assume the res])onsibility of providing clean seed from his crop foi' l)lanting extensively throughout the valley in 1914. In order to provide such clean seed he must take every precaution until the seed is finally sacked next fall to prevent mixture with other cotton. Planters of mixed seed by industrious attention to the removal of plants of foreign varieties from their mixed Durango fields can insure themselves a more valuable product in evener fiber than if they leave the short-staple plants to be picked with the Durango. By application to the matter of distinguishing the Durango plant type from plants of other varieties in his own field the planter of mixed seed takes a step toward learning to breed superior cotton. Cotton growers should compare Durango with Mebane plants at then* first opportunity and distinguish for themselves the differences in the leaf and bract characteristics-. No seed from mixed Durango fields should be preserved for i)lanting in 1914. Durango cotton 1 ^e t<> H inches long and of even-running length can be produced only by providing conditions of cidture that will allow normal, unaccelerated, and unchecked gi'owth of the plant. Under- watering or overwatering or any shocking of the plant growth ^^'ill tend to cause unevenness and other undeshable features in the fiber. Clean picking (to exclude trash) should be insisted upon, as it insures a clean grade, which is of decided advantage in selhng. The centralization of the ginning of all Durango cotton gl■o^\^l in the Imperial Valley will prove of definite advantage to gi'ower and ginner alike. ICir. IL'IJ THE CONTROL OF THE SUGAR-REET LEAF-SPOT.^ By V. W. Pool and M. R. McKay, Scientific Assistants, Cotton and Truck Disease and Sugar-Plant Investigations. INTRODUCTION. A detailed investigation of the leaf-spot of the sugar beet is at present bemg carried on in cooperation with the American Beet- Sugar Co., at Rocky Ford, Colo. Although the study is as yet incomplete, a brief account of the results thus far obtained and interpreted may prove helpful to those interested in sugar-beet problems. THE CAUSE OF THE LEAF-SPOT. The leaf -spot of the sugar beet is caused by a parasitic fungus. In its development this fungus, Cercospora heticola Sacc, produces fruit bodies or spores. Wlien these spores fall upon the surface of a beet leaf, the temperature and moisture conditions being favorable, they germmate and push through the breathmg pores mto the Ulterior of the leaf. After an entrance has thus been gamed the fungus continues to grow, kUls the adjacent leaf portions, breaks out through the dead area or ''spot," and here produces more spores, which in turn cause infections at other pomts on the leaf surface, and thus the life cycle of the fungus is continued. The masses of spores on the diseased areas give to the leaf-spot the grayish color which is so readUy recognized durmg July, August, and September. The spot is usually surrounded by a red ring, which represents the attempt of the beet plant to isolate the invading organism and pre- vent its further spread through the leaf. The individual spots do not increase to any great size, but new spots are contmually bemg formed, and when the number reaches approximately a thousand the entire leaf is killed and the beet plant is forced to produce more leaves in order to keep normal the manufacture of the sugar in the' plant. If too many leaves are killed the beet plant can only main- tain its normal sugar production by temporarily reducmg the amount in the root. Accordingly, sugar from the root is transferred and used in the development of new leaf tissue. Since this sometimes requires from 3 to 5 per cent of the sugar content of the root, the final total is correspondingly reduced. 1 Issued Apr. 12, 1913. [Cir. 121] • 13 14 CIRCULAR NO. 121, BUREAU OF PLANT INDUSTRY. THE OVERWINTERING OF THE ORGANISM. During 1912 search was made to fii\d, if possible, some plant other than the sugar beet which was harboring the leaf-spot organ- ism, and upon which a winter stage of the fungus might occur, but no such plant was found. There is still, however, the possibility of the existence of such a host. A detailed examination was also made of the beet balls of each variety of beet seed that the American Beet-Sugar Co. had on hand for the season of 1912 at Rocky Ford, Colo. This examination did not reveal any spores of the Cercospora fungus which were capable of germination. The work was carried on in a way which was posi- tive enough to admit of the assertion that no leaf-spot occurred m this vicinity durmg 1912 as a result of the beet seed used. How- ever, while Cercospora spores capable of germuiation did not appear upon the beet balls of the 1911 output, it can not be concluded that every year's output will give like results, smce at present there seems to be a possibility that the disease must be carried upon the seed into localities which are entirely isolated from sugar-beet growing sections. Recent reports of Cercospora heticola from Europe indicate that the spread of the disease from this source is slight. The soil as a harbormg agent of the leaf-spot fungus is of importance because of the manure, beet leaves, and crowns that are to be found in it. The possibility that manure carries the leaf-spot organism over from one season to the next is very slight, if, indeed, it can be considered to be a factor of any importance in the light of recent experunents. The indications are that the organism is entirely killed by its passage through the alimentary tract of cattle, pigs, and sheep. It is at least certam that its growth is greatly inhibited. Of course, where beet-top ''hay" is fed in such a manner that it is not entu'ely eaten up, this means of getthig rid of the disease would be at fault. It has been proved repeatedly during the last year, and thus far this year, that the leaf-spot organism remams alive upon the old beet tops, includmg the leaves, petioles, and crowns, which are left from the previous year's crop. It makes no difference where these tops are left; they are capable of producing infection. FIRST APPEARANCE OF INFECTION. Since, then, Cercospora is carried over from one year to the next upon the old beet tops wliich are left in the field and which later dis- integrate in the soil, the first ap})earance of the leaf-spot in the fol- lowing season should come on the first-formed leaves of the beet plants or those nearest the soil where the source of infection occurs. This is actually what ha{){)ens. In last season's work it was found to be the rule that in fields which were known to have had leaf-sj)ot |Cir. ll'll CONTEOL OF THE SUGAR-BEET LEAF-SPOT. 15 during the preceding year and where the tops were allowed to remain upon the ground, the first infection in the new crop appeared upon the leaves which rested upon the ground or were more or less sub- merged and held to the ground through irrigation. In contrast to this, a field of beets which followed alfalfa for the first time showed a period of infection about three weeks later than the preceding. Moreover, the first symptoms of the disease were upon the leaves which were well exposed to the air. Tliis may be explained in the following way: When the new beet leaves attain sufficient size the temperature and humidity being favorable, the fungous organism in the old leaves, which lie immediately beneath the new growth, becomes active and infects the new leaves near the ground. Now, if there happens to be no material affording such source of infection, as in the case of the second field mentioned, where sugar beets fol- lowed alfalfa, infection becomes impossible excej^t from living spores of the leaf-spot fungus carried in the air. Such infection occurs only after the spores formed upon the lowest leaves find their way into the air. They then fall upon the higher and more exposed leaves and produce infection here rather than upon the leaves near the ground, which are more or less covered. Thus, the second infection coming later and appearing on a different part of the plant points to the fact that if the first infection did not occur other infections would be impossible. CAUSES FOR THE SPREAD OF INFECTION. It has been proved to be a fact that the leaf spots increase only slightly in number upon the old leaves wliich have passed the period of their greatest activity and are beginning to yellow. Numerous spores, however, which are produced in these spots infect the sur- rounding green leaves, so that the source of infection is continued even after the old leaf is dead. When the condensed moisture upon the leaves dries, the spores of Cercospora are borne by the wind or simply float through the air to other leaves, or they may be carried from one plant to another by insects, by man, or upon various kinds of cultural implements. The irrigation water has been proved to be one way in which the organism may be carried from one place to another in the same field, or through the -drainage water to other fields. POSSIBLE CONTROL BY REMOVAL OF BEET TOPS. Since the beet tops and crowns are the great source of infection and afford a means of carrying the disease over from one season to the next, it is evident that care should be taken to remove these from the fields as thoroughly as possible. In order to do this, the tops should [Cir. 121] 16 CIRCULAR NO. 121, BUREAU OF fLANT INDUSTRY. be removed while still green. Otherwise, it" the material is allowed to remain upon the ground for two or three days some of it becomes extremely brittle and breaks up easily on account of the rapid drying which usually takes place. The methods of beet-top removal now in practice in the Arkansas Valley are at fault not only from the stand- point of the control of the leaf spot, but from an economic standpoint as well. Even where the beet tops of several rows are thrown to- gether, shocked, and later removed, the ground underneath these shocks is certain to be infected. The resultant beet-top hay, how- ever, is of good quality. This method of shocking holds the disease in check much better than the very general practice of leaving all material as it falls at harvest time. When the beet tops are left scat- tered indiscriminately over the ground, the leaves become so brittle that they break into small fragments, thus makhig it im})ossible to remove them from the field. If such a field is pastured by sheep, much of the diseased material is trampled into the ground and there results a poor control of the disease, a great loss of nutritive material, and a packing of the soil, all of which should be avoided. Even after all the beet tops are removed from the field there is still danger that in feeding them in a dried condition some of the material will become broken up and will not be eaten. These fragments become mixed with the manure and are returned to the land. It has been proved in recent experiments that these broken-leaf portions contain the Cercospora fungus in a viable condition and consequently are capable of producing infection. Thorough control of the leaf-spot fungus is assured if the green beet tops are made into silage. In the fall of 1912 a small silo, holding approximately 3 tons, was filled with green beet tops. A few layere of strav/ were added to absorb the excess moisture and to prevent extreme fermentation. I'ive pounds of salt were also added for each ton of material. In recent tests that have been made of the silage resulting therefrom, no fungus of any kind has been found to be alive. Numerous cultures examined for living Cercospora spores gave nega- tive results in all cases. A portion of the silage material was steril- ized and then inoculated with a fresh culture of the leaf-spot fungus, but no growth developed. This is in great contrast to the viability of the fungus out of doors, where directly after harvest it still has almost the same vitality and vigorous growth that it evinced several months before. It is very probable that the silo is to be the real solution of the leaf- spot problem. Some of the advantages from an economic standpoint to be gained by the proper utilization of the beet tops are given by L. wS. Ware in his book, "Cattle Feeding with Sugar Beets, Sugar, Molasses, and Sugar Beet Residuum," Philadelphia, 1902, pp. 93-96. ICir. 121] CONTROL OF THE SUGAR-BEET LEAF-SPOT. 17 Besides the roots proper, one may harvest a large quantity of leaves and tops. How- ever, there are many farmers sufficiently blind to overlook the precious quahties of these portions of the plant, and allow them to remain and rot on the field without rendering other ser\dce than that of supplying a portion of certain mineral elements * * * which have been absorbed by the root during its development. The money value of these leaves when used as a fertilizer is certainly less than that which would be derived fi-om feeding to cattle. Beet leaves and tops contain, it is true, a certain amount of salts which are useful to the soil, but on the other hand many of these mineral substances can be more advantageously utilized by feeding the leaves to cattle and collecting the manure; their fertilizing properties are not subsequently lost by the passage through the animal's body and during the interval the stock has been benefited by receiving a good, wholesome, green fodder at the very period of the year when it is most relished and is eaten with avidity. * * * Experience shows that the best results are obtained by feeding siloed beet leaves during early spring. * * * Many farmers allow sheep to run over their fields and eat the leaves during their passage. Under all circumstances such customs should be prohibited, as large quan- tities of leaves are necessarily trod under and are thus destroyed, which in reality means a waste, as far as their nutrient value is concerned. All average crop of leaves and tops is stated as being about 4.8 tons to the acre. CONCLUSIONS. (1) The leaf -spot of the sugar beet is caused by Cercospora hrticola Sacc, which lives through the winter upon the old beet tops of the precedmg season. (2) Results thus far obtained mdicate that the organism is unable to survive a passage through the alimentary tract of cattle, sheep, and pigs. However, it is impossible to prevent waste in feedmg the beet tops, and consequently all material is not eaten and may be returned m a viable condition in the manure to the land. (3) For a thorough control of the leaf-spot the beet tops should be removed from the fields while still green and should be made into silage. Cercospora beticola is killed when the beet tops are siloed. [Cir. 121] THE WORK OF THE HUNTLEY EXPERIMENT EARM IN 1912.^ By Dan Hansen, Farm Superintendent, Office of Western Irrigation Agriculture. INTRODUCTION. The work being carried on at the Huntley Experiment Farm (fig. 1) is devoted mainly to experiments wath crops under irrigation. These include crop rotation and tillage methods, variety testmg of field crops, and tests of fruit trees, small fruits, and vegetables. In addition to the work on irrigated land a tract of 20 acres \jing above the iiTigation canal is being used for a series of experiments in crop rotation and tillage methods under the supervision of the Office of Dry-Land Agriculture, and some additional work wdth field crops has been conducted on other tracts above the ditch. CONDITIONS ON THE PROJECT. In many respects the season of 1912 was rather unusual. Spring came late, and the cold, wet weather made it difficult to get land into proper condition for seeding. In early July an unusually severe storm, which came in the form of hail on some parts of the project, damaged gro^^•ing crops to some extent, but perhaps the greatest dam- age was done to the first cutting of alfalfa, both in the field and in the stack. The season's rainfall was somewhat above the average. A large proportion of the precipitation came during September and October and interfered seriously with the harvest of beets and other late crops. During the year 1912 the farms on the Huntley Project for which water applications had been made comprised a total irrigable area of 20,200 acres, included in 505 farms. Of tliis an area of 14,425 acres 1 Issued Apr. 12, 1913. The Huntley Experiment Farm is located on the Huntley (Mont.) Irrigation Project, adjacent to the Osljom town site. It comprises al)out 200 acres of public land withheld from entry by the Department of the Interior at the time of the opening of the project, to be used as an experiment farm. Of the 200 acres, only about SO acres are irrigable, some of the land being occupied by two railroads, the main irrigation canal, and a large waste ditch, and part of it lying above the canal. In addition to the land mentioned, a tract of 40 acres of the heavy land near the town of Worden is used for experiments in reclaiming alkaline soils. The work of the farm is inider the supervision of the OflBce of Western Irrigation Agriculture, Bureau of Plant Industry. Other ofTices in the Bureau of Plant Industry and the Montana Agricultural Experiment Station are cooperating in the investigational work. [Cir. U'l] 19 20 CIRCULAR NO. 121, BUREAU OF PLANT INDUSTHY. was actually irrigated, and harvests were reported from 12,742 acres. Alfalfa was planted on 1,.340 acres, but was not harvested. An area of 343 acres was not reported on. In Table I is summarized the more ALFALFA CROPS ELAL N 4) IRRIGA ROTA T I ED HUNTLEY EXPERIHENJ FARH ■Fig. 1.— Diagram showiiig the location of the held experiments at the Huntley Kxperimeiil Farm in 1912, not including the Worden tract. Kxperiments are conducted both above and below the main canal sliown at c. important statistical information of the different cro})s on the project in 1912. The figures were furnished by the United States Reclama- tion Service. rCir. 121] WORK OF THE HUNTLEY EXPERIMENT FARM IN 1012. 21 Table I.— Acreage, yields, and farm values of crops produced on the Huntley Reclamation Project in 1912. Crop. Alfalfa Native hay Com ." Grain Grain hay Potatoes. Sugar beets Truck crops Less duplications. Total Average value per acre Area (acres). 3,221 265 644 3.339 525 118 4,660 170 200 12, 742 Yields. Farm values. Unit of yield. Ton ...do ... Bushel. ..do ... Ton.-... Bushel. Ton Total. 8,930 427 4,830 74, 525 415 11, 125 38, 680 Aver- age. 2.60 1.61 7.50 n.M .79 94. 27 8.30 Ma.x- 5.70 2.00 40.00 1 132. 00 2.00 300. 00 18.00 Per unit of yield. $7.00 15.00 .50 .47 10.00 .50 6.00 Per acre. Total. $58, 730 6,405 2,414 35,236 4,150 5,562 232, 080 5,900 Aver- age. 350, 771 $18. 23 24.15 3. 75 10. 55 7.90 47. 13 49.80 34.71 Max- imum. $39.90 30. 00 20.00 2 42. 24 20.00 150.00 108. 00 27.50 1 Oats. - Oats at 32 cents a bushel. CLIMATIC CONDITIONS. The climatological observations were made in cooperation with the Biophysical Laboratory of the Bureau of Phiiit Industry. Table II summarizes the observations made during the years 1911 and 1912. Table II. — Summary of climatological observations made at the Huntley Experiment Farm during the years 1911 and 1912. Precipitation (Inches). Year. Character of data. Jan. Feb. Mar. Apr. May. June. July. Aug. Sept. Oct. Nov. Dec. Total. 1911 1912 0.64 .27 0.32 .21 0.00 .41 0.85 2.00 3.29 2.44 2.13 1 u 0.81 9 on 1.05 1.39 0.57 2.97 0.88 3.25 0.82 .75 0.13 .00 11.49 17.08 Evaporation (Inches). 1911 4.388 5.827 4.900 7.124 7.020 8.875 6.942 6.071 6.959 5.079 3.722 2.568 2.475 39.932 32.018 1912 Da ILY W IND Velocity (Miles pei i Hou r). 1911 Average 5. 6 5. 6 5.8 6.3 9.4 8.8 13. Oi 17.5 1 2.0 1.5 2.6 1.8 4.0 5.2 8.8 7.7 2.7 2.3 4.6 3.9 S.7 6.0 2.3 .6 4.0 3.7 7.2 6.5 2.1 .8 4.4 4.2 9.3 8.0 1.0 .9 4.2 5.6 10.0 14.7 1.3 1.5 5.4 4.2 11.6 9.7 1.4 1.0 5.5 7.8 11 5 1912 1911 do Maximum. . . 5.6 5.2 4.8 1912 1911 do Minimum. . . 12.8 10.8 12.1 14.6 1.5 2.7 1912 do .7 i.6 .9 Temperature (° F.). 24.9 38.7 55.0 69.0 -20.5 13 1911 1912 1911 1912 1911 1912 Monthly mean do Monthly maximum. do Monthly minimum. do 14.2 16.6 50.0 53.0 -26 -35 16.1 29.1 40.0 52.0 -19 - 5 39.1 18.7 74.0 62.0 - 3 -27 43.2 46.5 77.0 78.0 17.0 20.0 53.8 55.5 92.0 90.0 24.0 32.0 68.5 66.8 94.0 99.5 40.0 36.0 67.6 67.2 97.0 95.0 41.0 44.0 64.1 66.6 97.5 93.0 33.5 40.0 58.2 50.1 94.0 89.0 28.0 24.0 44.8 44.7 84.5 79.0 14.0 17.0 23.6 29.7 55.0 59.0 -26 1 [Cir. 121] 22 CIRCULAR NO. 121, BUREAU OF PLANT INDUSTRY. Table II. — Sunirnanj of cli mat oJog leal obsirraliovs made at the Jliintlttj Experiment Fann duriiuj the ijcurn 1911 and 191J — Continued. Killing Fko.sts. • Last in spring. First in autuinn. Length Year. Bate. Minimum tempera- ture. Date. Minimum tempera- ture. of frost- free period. 1911 May 26 May 13 ° F. 32 28 Sept. 18 Sept. 16 ° F. 28 31 Days. 114 1912 12.5 Fig. 2. — View in field K, where the irrigated rotations are conducted. There arc 2i» ditlerent rotations under test, including alfalfa, sugar beets, potatoes, wheat, oats, corn, and flax. EXPERIMENTS WITH IRRIGATED FIELD CROPS. CROP ROTATIONS UNDER IRRIGATION.^ For the })ur])ose of determining which systems of crop rotation are best suited to conditions on the i)roject, a field containing 70 one- fourth-acre plats is devoted to a series of two, three, four, and six year rotations (fig. 2), in which the following crops are grown in different sequences: Al*"alfa, sugar beets, potatoes, wheat, oats, corn, and flax. This cro]>rotation work will have to be continued for several years before results of much value can be obtained. The land on whicii these experiments are being conducted was j)laiited to oats in 1911, wliich may explain the rather low yields of some of the croi)s, })articu- larly the beets, it having been noted that beets following oats on this new ground do not as a rule give heavy yields. Table III gives the average and maximum yields obtained in the rotation ex])eriments in 1912. ' These experiments were under the immediate supervision of Mr. James M. Spain. [Cir. IL'I] WORK OF THE HUNTLEY EXPERIMENT FARM IN 1912. 23 Table III. — Average and viuximum yields obtained in the irrigated rotation expermient at Huntley in 1912. Crop. Alfalfa' Potatoes Sugar beets. Com Wheat Oats Flax Variety. Montana Burbank Klein wanzlebener. . . Minnesota No. 13 Pringle's Champion. Swedish Select Minnesota No. 25.... Num- ber of plats. 16 1.3 14 5 3 15 2 Yield per acre. Unit. Ton Bushel.. Ton Brshel.. ...do ...do ..do .Vver- 0.70 244.0 10.3 3(i. 4 25.3 48.7 14.6 Maxi- mum. 1.0 413.0 14.37 42.8 31.6 66.8 19.4 1 Planted May 10, 1912. METHODS AND TIME OF PLANTING ALFALFA. In 1911 an experiment in methods and time of planting alfalfa was started on 13 one-fourth-acre plats in field A-IV. The experiment in- cluded early and late plantings, planting early with w'heat as a nurse crop, and planting in rows 18 inches apart to be cultivated. The rate of seeding of alfalfa was 12 pounds per acre, except in the row plats, where it was 6 pounds per acre. In the nurse-crop plats wheat was planted at the rate of 1 bushel ])er acre. Table IV shows the yields obtained in 1911 and 1912 from planting by the different methods. Table IV.— Yields of alfalfa , vhcal hai/, andvheat on field A-IV at Huntley in 1911 and 191:J in alfalfa-planting experiment. Method and dale of planting. 3 plats planted May 5,1911 tons . 3 plats planted June 5, 1911 do..! 3 plats planted June 5, 1911, in 18-inch rows ..../...do..'. 2 plats planted May 5, 1911, with wheat as nurse crop, cut for hay do 2 plats planted May 5, 1911: /bushels., .tons.. With wheat as nurse crop, cut for wheat i"^ Yield per acre. 1911 (2 cuttings). 2.43 2.00 1.75 2. .34 46.7 1912 (3 cuttings). 5.64 5.35 4.98 5.40 4.93 Total (2 years). S.07 7.35 6. 73 2 5. 40 3 4.93 1 Wheat hay. Plus 2.34 tons of wheat hay. 3 Plus 46.7 bushels of wheat. It is seen that early planting has resulted in somewhat higher yields than have been obtained from late planting. The yields secured from the row planting are noticeably low. In comparing these results with those obtained by planting the alfalfa with a nurse crop of wheat, consideration must be given to the yields of wheat and of wheat hay, as shown in Table IV. It will be noted that the alfalfa yields in 1912 on the imrse-crop plats were but little lower than those obtained where the alfalfa was planted alone. [Cir. 121] 24 CIRCULAR NO. 121, BUREAU OF PLANT INDUSTRY. The experiment was repeated in 1912 on field A-III. The results are shown in Table V. Table V. — Yields of alfalfa and vhrat on field A-III at Huntley in 1912 in alfalfa- planting experiment. Method and date of planting. 3 plats planted Mav 11, 1912 tons.. 3 plats planted .Tmie 14, 1912 do 3 plats planted June 14, 1912, in i8-inch rows do 4 plats planted May 11, 1912: With wheat as nurse crop, cut for wheat bushels.. Yield per acre in 1912. First cutting. 1.46 .53 .24 Second cutting. 0.70 Total. 2.16 ..53 .24 44.2 Fig. 3.— Red clover on plat A-II-S, photographed June 13, 1912. This was planted in 1911, and yielded at the rate of 3.7S tons per acre in 1912. Table V shows that early planting resulted in higher yields than those from late planting, as was the case on field A-IV in both 1911 and 1912. The row planting yielded even less favorably on field A-III than it did on field A-IV. All the results so far obtained favor early planting. They oppose planting in rows where irrigation is practiced and where the crop is cut for hay. Further time is required to determine the fuial influence of the nurse crop. GRASS MIXTURES. While it is recognized that alfalfa is likely to remain the leading forage crop of the })roject, information is often desired as to the value of other forage plants, especially grasses, clovers, and grass mixtures. Such information will be ])articularly useful to farmers who expect to engage in dairying and wh<^ will therefore have need for cow pastures during a part of the j'car. In 1911 and 1912 an experiment in which grasses and clovers (fig. 3) were tested singly and in difl'erent combinations was conducted in cooperation with the {Cir, 121] WORK OF THE HUNTLEY EXPEEIMENT FAEM IN 1912. 25 Office of Forao^e-Crop Investigations, Bureau of Plant Industry. The results obtained during the two years are given in Table VI. Table VI. — Yields of grass mixtures on Field A-II. Huntley Experiment Farm, 1911 and 1912. Plat No. A-II-1 A-n-2 A-II-3 A-II-4 A-II-5 A-II-0 A-II-7 A-II 8 Crop or crop eomlmiation. Bromus inermis, orchard grass, tall oat-grass, meadow fescue, Italian rye-grass, timoth}-, redtop, Kentucky bluegrass, alfalfa, alsike clover, red clover, and white clover Brcimus inermis, orchard grass, tall oat-grass, Italian rye-grass, al- falfa, alsike clover, and white clover Brnmus inermis, orchard grass, Italian rye-grass, alfalfa, alsike clover, and white clover Bromus innmis, orchard grass, redtop, timothy, and alsike clover. . . Orchard grass, Italian rye-grass, slender wheat-grass, Kenti:cky blue- grass, alsike clover, red clover, and white clover Timothy and alfalfa Timothy and red clover Red clover Yield per acre (tons). 1911 1.31 .S6 .95 1.00 1.59 2.50 l.()5 1.69 1912 Total (2 years). 1 Plowed in the fall of 1911 be- cause of poor stand. 3.37 4.69 2.02 5.31 4.05 3.7S 4.32 5.69 3.61 7. SI 5.70 5.47 Plats A-II-1 and A-II-2 were planted on very heavy, low ground which was difficult to get into perfect condition for seeding, with the result that a poor stand was secured. These plats were therefore plowed up in the fall of 1911. Of the grasses in the combinations, Bromus inermis, orchard grass, and Italian rye-grass have made the best stand, Bromus inermis doing especially well. DISTANCE OF PLANTING AND THINNING SUGAR BEETS. Special work with sugar beets has l)een carried on in cooperation with the Office of Cotton and Truck Disease and Sugar-Plant Investi- gations. This included tests of time of irrigation and distance of planting and thinning. In the distance of planting and thinning experiment a series of 45 plats was used in 1912. Fifteen plats were planted in rows 18 inches apart, 15 plats in rows 20 inches apart, and 15 plats in rows 24 inches apart. These were thinned to 6, 9, 12, 15, and 18 inches apart in the rows. All plats were run in triplicate. The average yield from the entire series was 15.5 tons per acre. The results obtained indicate that, regardless of the distance of thinning, the 18-inch rows will give the best yields. The best combination of thinning and planting was the 18-inch rows thinned to 12 inches. The three plats planted in this way produced an average yield of 17.8 tons per acre, the highest yield obtained. There was but slight variation in the sugar content of the beets grown in the different plats, the average being 18.7 per cent. In the irrigation test w^ith beets it was impossible to follow the plan as outlined, because of the unusually copious rains during the season; therefore, the differences in the yields obtained in 1912 were not significant. [Cir. 121] 26 CIRCULAK XO. 121, BUREAU OF PLANT INDUSTRY. IRRIGATION TESTS AVITII FLAX. At the request of the Montana Agricultural Experiment Station, a series of 10 one-tenth-acre plats was devoted to an irrigation test with flax (fig. 4). The purpose was to determine the amount of water best suited to this crop and the proper time of applying water. The plan of tliis experiment was to give one, two, and three irrigations at different stages of plant growth on the different plats — in one case to give but one irrigation, and that before planting. The unusual weather conditions, especially the heavy rain in early July, tended to bring on all plats uniformly, so that the results obtained would not apply to the average season. It was possible, however, to test the effect of late irrigation. Plats irrigated after the flax was well through the bloom did not show any tendency to start a second growth and were not held back in ripening, all plats being ready for harvest on about the same date. It mav be, however, that this would Fir,. 4.— Plats of Hax in field All, used in ihe experiment with time of irrigation. The average yield of the 10 plats in this e.xperiment in 1912 was 18.6 bushels per acre. not appl} to a large field, especially one that was not perfectly drained and where the water was permitted to stand for any length of time. The average yield of aU the plats was at the rate of 18.6 bushels per acre, and variations from this were so sUght that they could not pro})erly be attributed to differences in treatment. FIELD CORN. A test to determine the earliest and best-yielding varieties of corn suitable for use on the project ami in the rotation experiments was car- ried on during 1912 in cooperation with the Office of Corn Investiga- tions. The varieties yielded as follows: Northwestern Dent, 58.3 bushels; Brown County YeUow, 51.5; Wisconsin No. 7, 61.8; ^linne- sota No. 13, 59; Minnesota No. 23, 51.7; and Selection No. 133, 62.9 bushels per acre. The average yield of all the varieties was 57.5 bushels per acre. The fh'st three varieties matured better than the others. On account of their earfiness they are more promising for the Huntley Project. The Northwestern Dent and the Brown County Yellow were tlie best varieties tested in 1912. [Cir. 1-lJ WORK or THE HUNTLEY EXPERIMENT FARM IN 1912. 27 ORCHARD AND SMALL-FRUIT TREES. In the spring of 1911 a 5-acre tract was planted to about 100 varieties of apples, cherries, and plums, together with 26 varieties of small fruits.^ The unusually severe weather of the ^vinte^ of 1911-12 mjured many of the trees, especially apples and cherries, and it was necessary to replant about half the trees in the spring of 1912. All trees made good growth during the season. Small fruits came through the winter in good condition. RECLAMATION OF THE WORDEN TRACT. Some of the land in the Huntley Project is heavy clay, carrying more salt than is tolerated by most crop plants. This salty condi- tion is due chiefly to the impervious character of the soil and the Fig. 5.— Cultivating plats in field 11. AVorden tract, in the test of alternate irrigation and cultivation. The salt content of the surface 12 inches of soil has been reduced in one year from an average of 0.52 per cent to 0.2S per cent. consequent lack of leachmg by the rains. The subsoil at a depth of about 6 feet is a porous gravel, and the chief problem is to get the surface soil opened up and supersaturated with water to bring about artificial leaching. In 1910, 12 acres of a 40-acre tract of typical heavy land near the Worden town site were broken up, and experiments have been carried on there to determine the best means of reducmg the salt content and bringing the land into production. One method employed was to level the ground m plats of about one-fourth of an acre each, around which a border was made, and to practice on these plats a system of alternate Ught irrigation and cultivation (fig. 5). This work has been carried on but one full 1 These varieties were selected upon the advice of Mr. H. P. (lould, of the OfTice of Field Investigations in Pomology. [Cir. 121] 28 CIRCULAR NO, 121, BUREAU OF PLANT INDUSTRY. season, but salt determinations made at different times show a decided reduction in the salt content of the upper 12 inches of soil. The average salt content of the upper 12 inches of soil on plats M-I-5 to M-I-9, mclusive, in September, 1911, was 0.52 per cent. By September, 1912, the percentage of total salts in the same plats had been reduced to an average of 0.28. About 10 acres of the land have been planted to wmter lye for two seasons. Both crops of rye were produced without irrigation, the winter rains and snows giving sufficient moisture. The crop was plowed under m May each year and the land kept in good tilth throughout the remainder of the season in order to prevent the evaporation of water from the surface and the consequent accumu- lation of salt in the upper layer of the soil. Fig. (i. -A plat of rye on the Wonieii tract in May, 1912. Rye was grown here in 1911 and 1912, the .sefond crop l:)einK heavier and more uniform than the first. The green crop has been plowed under each year to improve the physical condition of the surface soil. Salt determmations have not been systematically made on the rye land, so that it is not known to what extent the salt content of the surface soil has been reduced. The growth of rye (fig. 6) was much heavier and more uniform in 1912 than it was m 1911, and the soil tOth was decidedly better, so that the plowing under of the rye crop appears to have had a beneficial effect. Durmg the season of 1913 salt determinations will be made regularly and the salt content of the rye land will be compared with that of the adjacent virgin soil. [Cir. 121] METHODS OF SECURING SELF-POLLINATION IN COTTON/ By Rowland M. Meade, Scientific Assistant, Office of Crop Acclimatization and Adapta- tion Investigations. The advantage of being able to prevent cross-pollination and thus insure pure seed is recognized by all cotton breeders. Various meth- ods have been used, both in this country and abroad. Some experi- menters liave gone so far as to cover single plants with cages of fine- mesh wire or cloth in order to prevent the visits of insects to the flowers. This method may be very convenient when pure seed is desired from a single plant only, but is scarcely practicable for whole rows or bulk plantings. The expense of covering the plants is great, and if the mesh of the screen is fine enough to exclude thrips and other very small insects the conditions of growth are affected, so that the plants may not develop normally. The method of bagging the flowers has been used most extensively. A small paper bag is mflated and placed over a flower bud and the mouth of the sack is tied with string or wire about the pedicel. Flowers covered in this manner open naturally inside the sacks, and the pollen is distributed from the anthers as freely as in uncovered flowers. Within a few days after fertilization has been accomplished, the paper sacks are again removed to allow normal development of the young boh. The first operation of adjusting the sacks is one that requires some time and skill, while the operation of removing the sacks greatly uicreases the labor and requnes additional time. Another method has been described by Mr. W. W. Gilbert, of the Bureau of Plant Industry, in the Proceedings of the American Breed- ers' Association, vol. 8, 1912. Briefly, it consists of coiling a fine, soft copper wire in a spiral about the enlargmg flower bud, making one end fast near the base of the petals and bendmg the other end double over the tip of the bud. This coil keeps the petals from open- ing and the doubled end prevents the entrance of insects. The wire falls away with the petals, which avoids a second operation like that of removing the bags. The work of coiling and adjusting the wires can also be saved by using other devices for holding the petals together. Dr. D. A. Saun- ders, of this Bureau, has found it possible to use small rubber bands instead of the coiled wires. The band is snapped over the upper end of the flower bud and does not aflow the petals to open, but permits sufficient expansion of the petals for the normal development of the 1 Issued Apr. 12, 1913. [Cir. 121] *^ 29 30 CIRCULAR NO, 121, BUREAU OF PLAINT INDUSTRY, stamens and anthers. A great many flowers may bo treated in this manner in a short time, for only a few seconds are required to sna]) the bands over the buds. An equally simple and efficient method employs a familiar paper clip of the form in common office use. The sides of the clip arc separated and the ends spread out as in figure 1, A. This is then placed over the end of the bud so that the tip of the bud projects through the loop of the clip and is held by a free end and one side of the clip. (See fig. 1,5.) The petals can expand to a cer- tain extent, but the clip holds them together sufficiently to prevent their opening. Enough space is provided for the normal development of the stamens, and pollination is assured. This method has an advantage over the rub- ber-band scheme in its man- ner of securing the petals. The clip holds the petals lengthwise and does not slip off, while the rubber band secures the petals laterally and is sometimes forced off by the expandmg bud. Further modifications of these methods may be possi- ble without changing results. Any appliance that can be ])laced over the bud to pre- vent its opening and still allow sufficient expansion of the petals for the normal development of the stamens and pistil will provide all the require- ments that insure self-fertilization in the cotton flower. Any of these methods may also be used to prevent the openmg of a flower from which pollen is to be taken in hybridizing work. The advantages of these simplified methods will be most readily appreciated by breeders who wish to maintam a pure line of cotton, as they make it possible to s(H-ure a larger quantity of pure seed in a limited time. Fig. 1.~(A ) Wire paper clip adjusted for placing on cotton bud. (B) Unopened flower budwithwireclip in position to prevent the separation of the petals. (Natural size.) ICir. ll'lj o U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— Circular No. 122. WILLIAM A. TAYLOR, Chief of Bureau. MISCELLANEOUS PAPERS. Directions for Blueberry Culture FREDERICK V. CyVILLE The Work of the Truckee-Carson Experiment Farm in 1912 Feterita, a New Variety of Sorghum F. B. HEADLEY I H. N. VINALL - 1 and CARLETON R. BALL Issued April iq, igij. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1913. BUREAU OF PLANT INDUSTRY. Chief of liuicuu, William A. Taylok. Assifstunt Chief of Bureau, L. ('. Ooubett. Editor, J. E. Kockwkll. Chief Clerk, James E. Joxes. [Oir. l-J-_'] [Cir. 12J— A.] DIRECTIONS FOR BLUEBERRY CULTURE/ By Frederick V. Coville, Botanist in ClKinjv of Ta.roiioinic ~>° F. if practicable. When sub- jected to high temperatures the newly cut shoots are liable to die and rot from the base upward. The outer surface of the pots should never be allowed to become dry. The desired condition may be assured by bedding the pots in moist sand up to the rim. 8. Watering should be as infrequent as practicable, only sufficient to keep the soil moist but well aerated. 9. The frame should receive ventilation, but not enough to cause the new twigs to droop. These are most susceptible to overventilation and to overheat- ing when they have nearly completed their growth. 10. After the new twigs have stopi)ed growing and their wood becomes hard, new root growth takes place. Then secondary twig growth follows, either from the apex of the new twigs or from another bud lower down on the old wood of the original rooted shoot. Until this secondary twig growth takes place the life of the plant is not assured. 11. Those plants that make sulhcient growth to require repotting during the first summer should be set in clean pots of 2 inches larger diameter in a standard blueberry soil mixture. SOIL MIXTURE FOR BLUEBERRIES. A very successful potting mixture, or nursery-bed mixture, for blue- berry plants consists of one part of clean or washed sand, nine parts [Cir. 122] 6 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. of rotted u])l;iii(l peat, cither eliopped or riiblKHl throiij>h a sieve, and three i)arts of clean broken crocks. No loam, and especially no lime, should be used. Manure is not necessary, and in the present state of our knowledge may be regarded as dangerous, although in small amounts it serves to stimulate the plants, at least temporarily. The danger from manure apparently lies in its tendency to produce an alkaline condition in the soil. The use of broken crocks in the potting mixture is based on the fact that the rootlets seek them and form around them the same kind of mats that they form at the wall of the pot, thus increasing the effective root surface and the vigor of growth. The peat most successfully used for potting blueberry plants is an upland peat procured in kalmia, or laurel, thickets. In a sandy soil in which the leaves of these bushes and of the oak trees with which they usually grow have accumulated and rotted for many years un- touched by fire, a mass of rich leaf peat is formed, interlaced by the superficial rootlets of the oak and laurel into tough mats or turfs, commonly 2 to 4 inches in thickness. These turfs, ripped from the soil and rotted from two to six months in a moist but well-aerated stack, make an ideal blueberry peat. A good substitute is found in similar turfs formed in sandy oak woods having an underbrush of ericaceous plants other than laurel. Oak leaves raked, stacked, and rotted for about 18 months without lime or manure are also good. The leaves of some trees, such as maples, rot so rapidly that within a year they may have passed from the acid condition necessary for the formation of good peat to the alkaline stage of decomposition, which is fatal to blueberry plants. Even oak leaves rotted for several years become alkaline if they are protected from the addition of new leaves bearing fresh charges of acidity. TUBERING. By ordinary methods, cuttings of the swamp blueberry have been rooted only in occasional instances. Two successful methods, how- ever, have been especially devised for these plants. The most novel of these but the one easiest of operation is that of " tubering." This method involves the same principle as that employed in stumping, namely the forcing of new shoots in such a manner that their basal portions are morphologically scaly rootstocks, with a strong rooting tendency. The directions for tubering as applied to the swamp blue- berrv are as follows: 1. Make stem cuttings from ontdoor plants between midwinter and early spring, before the Ituds have begun to make their spring growth, and prefer- ably on a warm day wlien the twigs are not frozen. A still better i)lan is to make Ihe cultings in autuuni after tho leaves have fallen, and store them for [Cir. 12LM DIRECTIONS FOR BLUEBERRY CULTURE. 7 .-ibout two mouths in moist si)lj;i.i;iium on iro al a tempera I luc Just above freezing. 2. The cuttings are to be made from vigorous plants grown in well-lighted situations and with stems therefore well stored with starch. T'se unbranched portions of the old and hardened branches and stems, about a quarter of an inch to an inch or e\en more in diameter. Three to four inches is a suitable and convenient length. Make the cuts with pruning shears or a fine-toothed saw and remove the bruised wood at the cut ends with a sharp knife. Be cai'eful not to injure the bark or split or strain the wood. 3. Lay the cuttings horizontally in a shallow box or other cutting bed of pure clean Siind and cover them to the depth of about half an inch. :Moisten the sand well with rain water, bog water, or other pure water (free from lime) from a sprinkling pot. and see that the sand is closely and firmly packed about the cuttings. Cover the box or cutting bed with a pane or panes of glass, the top of the box being flat, so that the glass fits it rather tightly. The i)ox should be so prepared that any surplus water in the sand will drain away beneath through holes in the bottom covered with clean broken crocks and sphagnum moss. 4. Keep the box at a temperature of 55° to 65° F., or as near those limits as practicable. A temperature of 70° or over is likely to ruin the cuttings. 5. In order to avoid excessive temperatures, do not allow direct sunlight upon the glass, either keei>ing the box by north light or keeping it shaded, as by a white cloth or palmer cover susi^ended several inches above the glass or in a shaded greenhouse. 6. Keep the air inside the box saturated with moisture. This condition will be evidenced by the condensation of the moisture on the under side of the glass during the cooler part of the day or whenever a cold wind blows against the glass. 7. Watering should be as infrequent as practicable, only suHicient to keep the sand moist but well aerated and the atmosphere in the box saturated. If the glass fits tightly, a second watering may not be needetl for several weeks, when ventilation of the growing shoots has reduced the moisture of the cutting bed. 8. Within a few weeks new growth will begin to appear above the sand. Ventilation may then be given by sliding the glass until a crack is left open at one side of the box, but the air beneath the glass should still remain in a condition of approximate saturation. If the new growth starts to wilt, reduce the ventilation till the wilting ceases. 9. When the shoots have reached a length nroixtrtionate to their vigor, com- monly 1 to 3 inches, their further growth is self-terminated by the death of the tip. After the leaves have reached their full size and acquired the dark-gi-een color of maturity the time has come for the development of roots. 10. When the first shoot has reache vears old and about 1^ to 2 feet high. They are best set out in early spring, before the buds have begun to push. [Cir. 122] DIRECTIONS FOR BLUEBERRY CULTURE. 9 It is a curious fact that these plants send out no new i-oots in spring until the}^ are in full leaf, their flowering is nearly or quite finished, and their principal twig growth has ceased. It is important, therefore, in taking up either a wild or a cultivated plant from the open gi-ound that as much as possible of the old root mat be lifted with the plant, for upon this they depend for moisture until their new rootlets are formed. In the case of mature wild bushes with very large root systems, when it is practicable to secure but a fraction of the root mat, say. a disk only 3 or 4 feet in diameter, it is the best procedure to cut all the stems to the ground at the time of transplanting. The bush will then produce a new and symmetrical top of a size suited to the capacity of the roots. The wood that is removed may be used for cuttings if the plant is sufficiently valuable. In the permanent field plantation the bushes should be set 8 feet apart each way. When they reach mature size they will nearly or quite cover the intervening spaces. To secure full vigor of growth the ground between the bushes must be kept free from all other vegetation. On roclr^^ uplands a continu- ous mulch of oak leaves, Avhen it is practicable to secure them, will help toAvard this end as well as keep the soil in the necessary acid condition. It is more economical, however, to choose such a location for the plantation as will permit the use of horse-drawn machinery and will make mulching unnecessary. The most favorable location for blueberry culture is a boggy area with a peat covering and sand subsoil, the peat preferably of such a ihickness that a deep plowing Avill turn up some of the underlying sand. The land should be so ditched that the water level can be kept at least a foot below the surface of the ground during the growing season or can be raised for subirrigation during a drought. The ground should be plowed to the depth of about 8 inches and repeatedly harrowed or otherwise tilled during the season preceding the planting, in order to kill the vegetation. The best time for plow- ing is late spring, after the principal vegetation has used up its stored starch in completing its early growth and before the leaves have matured and the roots have begun the new storage of starch with uhich they can send up new sprouts. After the plants are old enough to have formed a root mat, the tillage should be very shallow, not more than 2 or 3 inches, so that the roots will not be injured. This is probably best accomplished by the use of a light spring-tooth harrow with the teeth set closer together than usual. Fertilizer experiments have .shown that lime is positively injurious and that manure, while producing a temporary stimulation of vege- 87087°— Cir. 122— 13 2 10 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. tative growth, is likely to cause serious injury later. For those desir- ing to experiment with fertilizers the following acid mixture is rec- ommended, applied at the rate of 1,000 pounds per acre, or one-fifth of a pound per square yard : Pounds. Acid iibosiiliate (liigli grade, about 36 per cent available pbosphoric acid) — 600 Sulpbate of potasb (50 per cent potash) 200 Sulphate of ammonia (20 per cent nitrogen) 200 (Muriate of potash may be substituted for sulphate of potash.) This and similar acid mixtures have been used with success on blue- berry plants, in both pot and field experiments, wdth no evidence thus far of cumulative injurious effects. However, as no fertilizer is re- quired to make the swamp blueberry fruit abundantly and continu- ously in suitable peat and sand soils properly handled, the use of fer- tilizers in commercial plantations is not at present advocated. The swamp blueberry does not require a yearly pruning. A^^ien one of the stems of a bush becomes unproductive from injury or old age it should, of course, be cut out. If a large part of a bush needs removal it is better to cut all the stems to the ground and let the plant send up new shoots all of the same age to form a Avholly new and symmetrical top. YIELD AND PROFITS. By proper manipulation in the greenhouse, seedling blueberry plants can often be made to ripen a few^ berries in less than a year, but they do not come into commercial bearing in field plantations until they are about 5 years old, when the plants are 3 to 4 feet high. They then increase slowly to full size and full bearing. Wild bushes of the swamp blueberry live to great age, often 50 to 100 years, still bearing heavily. Individual stems may remain productive for 10 to 25 years. When dead they are replaced by new and vigorous shoots from the root. The field plantings resulting from the recent experiments in blue- berry culture are too young to show the mature yield. Fortunately, however, there has been found, near P^lkhart, Ind., a small blueberry planting of mature age, believed to be the only commercial plantation in existence. The returns from this plantation set forward our knowdedge of yields by at least a decade. The plantation is a little less than 2^ acres in extent. It was started in 1889 in a natural blue- berry bog, wdiich was first drained and then set with unselected wild l>luel)erry bushes. The plantation was profitable from the first, but exact records of yield and receipts are available only for the years tCir. 121.'] DIRECTIONS FOR BLUEBERRY CULTURE. n 1910 to 1912, when the phmtation was 21 to 23 years old. The data are as follows : Yield and receipts from i)lantation of hluehcrries near Elkhart, Ind., 1910 to 1912. Year. Yield per acre. Price (ap- pro.ximate average per quart). Receipts per acre. Profits per acre. 191(1 (a j-ear of "almost total failure" because of late spring freezes) Quarts. 419 2, 26(1 2,379 Cents. 121 S71.87 292. 44 305. 75 $10 139 147 1911 1912 The annual expenses for weeding, cultivation, and irrigation were about $20 per acre. The cost of jjicking was 5 cents a quart. The general cost of maintenance of the equipment w^as about $2 per acre per year, the crates and boxes being used repeatedly. The computa- tion includes an estimated annual charge of $12 per acre for interest, $2 for taxes, and $4 for depreciation or sinking fund. It must be borne in mind that these figures are based on the yields from w^ild bushes transplanted without selection as to individual productiveness or the size of the berries. With bushes propagated from selected .stocks the yield should be greater and the berries much larger, this greater size probably effecting a reduction in the cost of picking and certainly an increase in the selling price. Onl}^ a beginning has been made in the improvement of the blue- berry. In a series of experiments involving the selection of superior Avild strains, the growing of hybrids, and the forcing of choice varieties to quick fruiting by budding them on strong seedling stocks, berries eleven-sixteenths of an inch in diameter have already been produced, and it is expected that from the better Avilcl stocks now available berries of still larger size will be developed. [Cir. 12:>J [fir. 122— B.] THE WORK OF THE TRUCKEE-CARSON EXPERIMENT FARM IN 1912/ By F. B. Headley, Farm f^iiitcrintoitlent. Office of Western Irrigation Af/riciilturc. INTRODUCTION. The Tnickee-Carson Experiment Farm - was established by the Bureau of Plant Industrj^ of the Department of Agriculture in 1906 on 160 acres of land 1 mile south of Fallon, Nev. The land was with- drawn from entry by the Reclamation Service for use as an experi- ment farm. The Truckee-Carson Project is situated in one of the driest regions in the United States, the average annual rainfall being less than 5 inches. The native vegetation of the project is extremely sparse, so that the virgin soil is lacking in vegetable matter. Furthermore, the high evaporation has resulted in the accumulation of alkali salts in the surface soil on many parts of the project. Since irrigation was commenced several years ago, surplus water has accumulated in many of the low-lying lands and a water table has thus been established. In certain parts of the project this water table is within 2 to 4 feet of the surface and its presence has a detri- 1 Issued Apr. 19, 1913. = The Truckoe-Carson Experiment Farm is maintained by the Office of Western Irriga- tion Agriculture of the Bureau of I'lant Industry. For a report on the worli of the farm previous to 1909, see Scofleld, C. S., and Rogers, S. J., " Tlie Truclcee-Carson Experiment Farm," U. S. Department of AgricuUure, Bureau of Plant Industi-y, Bulletin l."iT, 1909. More recent publications relating to the investigational work are as follows : U. S. Department of Agriculture, Bureau of Plant Industry : Bulletin 211, 1911. Bacteriological studies of the soils of the Ti-uckee-Carson Irriga- tion Project, by K. F. Kellerman and E. R. Allen. Circular 91, 1912. The nematode gallworm on potatoes and other crop plants in Nevada, by C. S. Scofleld. Circular 110, p. 21-25, 1913. Agriculture on the Truckee-Carson Project: Vegetables for the home garden, by F. B. Headley and Vincent Fulkerson. Circular 113, p. 15-22, 1913. Commercial truck crops on the Trujckee-Carson Project, by F. B. Headley and Vincent Fulkerson. Circular 114, p. 25-30, 1913. Climatic conditions on the Truckee-Carson Project, by F. B. Headley. Circular 118, p. 17-2S, 1913. Fruit growing on the Truckee-Carson Project, by F. B. Headley and Vincent Fulkerson. The arrangement of the fields and the location of the crop experiments in 1912 are shown in figure 1. 13 [Cir. 122] 14 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. mental effect on the soil and on crops. The three conditions above enumerated, namely, the general lack of vegetable matter in the soil, the presence of harmful quantities of alkali salts, and the existence of a high water table, determine the nature of most of tlie problems of crop production on the j^roject. The investigations conducted at the experiment farm have been concerned largely with these three conditions. (Fig. 1.) One of the first requisites of successful crop production on the project is to get the soil well supplied Avith vegetable matter, and the /I L F o W TRUCKETE-CARSON CXPERIMENT FAR PI 6 Fio. 1. — Diagram of the Trufkee-Carson ExptTiiin'iit Farm, sliuwiiii;- the arrangcmi'iit of the fiplds and the locatio!i of the crop experiments in 1!)1L'. most practical method of accomplishing this is to get a crop of alfalfa established on the land. This has been one of the principal aims of the work at the experiment farm. Numerous experiments have been conducted with a view to reducing the salt content of the soil, and in 1912 a i^umping plant was installed for the purpose of lowering the level of the ground water. The question of marketing crops is also an important one to the farmers on the project. The local market demands are so limited It'ir. \22\ WORK OF TRUCKEE-CARSON EXPERIMENT FARM IN 1912. 15 that it will be necessary for the farmers to produce crops which can be profitably shipped to distant points. Commercial truck crops, such as potatoes, onions, and melons, are very promising in this connec- tion, and much of the work of the experiment farm has been devoted to the testing of varieties and cultural methods for these crops. Some work of an educational nature is done among the local farm- ers. Instruction and demon.strations are given in connection with the proper methods of tree planting, pruning, and spraying, and an active part is taken in the local agricultural fairs. Some attention has been given to finding what shade and ornamental trees and other plants could best be used in beautifying home grounds. {Fig. 2.) CLIMATIC RECORDS. Daily records are kept of the maximum and minimum temperatures, humidity, precipitation, evaporation, and wind velocity. This work is carried on in cooperation with the Biophysical Laboratory of the Fk;. 2. — A hedgv of oleaster on the Truckee-Carsoa Experiment F.inii. For ornamental purposes and as a \viudt)reak oleaster is one of tbe most desirable plants tested at the farm. Bureau of Plant Industry. A meteorograph has been installed on the experiment farm by the University of Nevada. This instrument records automatically on a single sheet of paper the relative humidit}^, temperature, barometric pressure, wind velocity, and wind direction. Table I gives a summary of the climatological observations made since 190G. Table I. — Hinniiiai!/ of cliiiKitolofficdJ oJ)scn-(it!oiis er acre. Sodium nitrate, potassium sulphate, acid phosphate, and gj-psum applied at the rate of 7.'"tO pounds per acre. EFFECT OF ALFALFA ON STTGAR BEETS. Several fields on private ranches were surveyed for the purpose of determining the yields of sugar beets on various types of land. The results are shown in Table III. Table ITT. — Effort of rarioiis tiiprfi of hind on tlic i/irltis of siif/nr hectic. Previou.s crop. Alfalfa. None. . Kind of soil. Loam (Adobe \Sandy loam. Area. Yield per acre. A cres. Tom. \ 21.3 19.4 \ 1.16 25.3 1 4.3 18.9 2.26 3.6 2.46 9.9 Sugar content. Per cent. 18.1 18.0 18.8 21.8 19.8 These results indicate that a previous growth of alfalfa has a decidedlv beneficial effect on the yield of sugar beets. SWEET C'IX»VER. Sweet clover has always been considered as undesirable in many sections of the country, but farmers on the Truckee-Carson Project have come to think well of it. It has been grown for a number of years at the experiment farm, usually on soils where alfalfa would [Cir. 122] WORK OF TRUCKEE-CARSON EXPERIMENT FARM IN 1912. 19 not grow well. It can be grown successfully on soil containing more black alkali than is tolerated by alfalfa. It has been found that if the crop is cut before blossoming, live stock soon learn to eat it. It has been observed that some horses and cattle eat it as readily as they do alfalfa, while it is apparently disliked at first by others. It is not recommended that this crop be grown on land that will grow alfalfa successfully, but it may frequently be used on land which for some reason will not produce other crops. The sweet clover may be used as a forage crop to feed live stock or it may be plowed under Avlien several feet high to help bring the soil into better condition. Fig. 3. — Row tests of lUfalfa varieties at tlie ■rrucljee-Carson Experiment Farm in 1JJ12. Twelve varieties were tested in tliis manner. When sweet clover comes up in an alfalfa field it can be killed out the second year, provided the crop is cut during the two years early enough to prevent the clover from producing seed. ALFALFA. In April, 1911, 12 varieties of alfalfa were seeded in rows 30 inches apart. (Fig. 3.) The row^s were from 320 to 330 feet in length. The entire series w^as planted in duplicate. The seed of these varieties was furnished by the Office of Forage-Crop Investi- gations. The results obtained in 1912 are presented in Table IV. [Cir. 121'] 20 CIRCULAR NO. 122, BUREAU OP PLANT INDUSTRY. Table IV. — Yields of alfalfa varieties in row lests at the Tructcee-Carson Ex- periment Farm in 1912. Variety. Cauca.sus Medicago ruthenica Arabian Peruvian Grimm Sand lucern Turkestan Montana Canadian Western Grown Provence Elche First Second crop. crop. Pounds. Pounds. ZJO 135 SO 95 140 140 215 155 115 115 1S5 120 185 120 195 1.35 175 145 1.50 120 S5 120 Tiiird crop. Total. Pounds. Pounds. 38 1 82 : 91 57 43 49 69 59 70 53 02 403 257 .371 427 273 354 374 389 390 323 267 Length of row. Feel. .320 322 324 325i 326" 327 328 330i 333 334,1 335 336 Yield per KW feet of row. Pounds. 126 79 114 131 83. i 108 113 117 117 96 79 The Grimm, Caucasus, Western Grown, and Canadian varieties gave the largest yields. The Grimm alfalfa started early in the spring and was ready to cut before any of the other varieties. MILLET. Four varieties of millet were grown, namely, the Jai^anese, Colo- rado Golden, Hog, and Xew Siberian. Of these the Japanese jdelded the greatest amount of hay, but the Colorado Golden was the earliest and set a large quantity of heavy, plump seeds. The Colorado Golden promises to be valuable on account of its short season and high seed j^roduction. Further tests will be made with it for the purpose of ascertaining its seed yield per acre. riEIJ) CORN. A number of varieties of field corn have been tried for several years without obtaining at any time sufficiently large yields to indi- cate that corn will be a commercially profitable crop in this section, except possibly in the most fertile soils of the old river bottoms. During some years the early dent varieties have a sufficiently long period to mature, but it is probable that for the average season some of the flint varieties will give the highest yields. The Australian White Flint was the earliest and best yielding variety tried in 1912. HORTICULTURAL WORK. There are two small orchards on the experiment farm. One of these came into bearing for the first time in 1912. Observations have been made on the bearing orchards on some of the older ranches of the project during the past three years. Identifications have been made of many of the varieties and notes as to their relative ])roduc- tivity have been taken. A few of these old orchards have been [Cir. 1221 WORK OF TRUCKEE-CAESON EXPERIMENT FARM IN 1912. 21 I^lotted and. so far as possible, the variety name of each tree inserted. The results of the horticultural Avork have been published separately.^ TESTS OF VEGETABLES. For several years variety tests have been made with a number of vegetables. Good results have been obtained with asparagus, beans, table beets, carrots, sweet corn, cucumbers, kohl-rabi, melons, onions, peas, potatoes, pumpkins, rutabagas, squashes, tomatoes, and turnips. The results of this work up to and including the season of 1912 have recently been published in detail, together with directions applicable to each crop.^ COOPERATIVE WORK WITH FARMERS. Since the soil of the experiment farm is not yet in suitable condi- tion for experimental crop work, it has seemed advisable to conduct some of the more urgent crop tests on adjacent older farms, where the soil is better adapted to this work. The greater part of the coop- erative work in 1912 was done with potatoes, onions, and strawberries. POTATOES. Arrangements were made in 1912 with J. W. Ferguson, Fallon district; W. Vanvoorhis, Stillwater district; Dr. Richards, Island district; and J. W. Rawles, Fernley district, for conducting experi- ments with fertilizers on potatoes. Seed potatoes from a northern State Avere used, to avoid danger of introducing eelworms. A variety known as Colorado Pearl, or ^Vh\te Pearl, was purchased from a seed firm in Utah. It was found, after planting, that the germination of this seed was poor, and consequently a poor stand resulted on all the farms, so that at harvest time the fertilizer effect could not be determined. ONIONS. A piece of land was selected on the farm of A. E. Merritt as suit- able for the onion test. The land had previously been in clover pasture for three years and was manured with well-rotted stable manure during the winter prior to planting. One hundred and ten rows of onions were planted, with varying quantities and kinds of fertilizers. At the end of the season each row was harvested and weighed separately. It was found that the 1 U. S. Department of Agr/cuUni-e, Bureau of Plant Industry, Circular 118, p. 17-28, 1913. Fruit growing on the Truckee-Carson Project, by F. B. Headley and Vincent Fulkerson. 2 Bureau of Plant Industry, Circular IIM. p. UI-lT), l!ti:{. Agriculture on the Truckee- Carson Project: Vegetables for the home garden; and Circular ll."!, p. 13-2L', 1913. Commercial truck crops on the Trucke«-Carson I'roject. [Cir. 122] 22 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. average difference between the yields of unfertilized and fertilized rows was insignificant. Tlie indications are that on rich soil, such as was used in this experiment, commercial fertilizers are of little or no value. STRAAVBERRIES. In April, 1912, 25 plants each of 20 varieties of strawberries were ]:)lanted on the farms of W. J. Ferguson, P'allon district ; W. Van- voorhis, Stillwater district ; and W. W. Cogswell, Fernley district. As strawberries do not produce a crop until the second year, a report can not be made at the present time on the relative desirability of these varieties. ALKALI AND GROUND- WATER STUDIES. Since the establishment of the experiment farm studies have been made to determine the causes of the poor growth made by plants on much of the land, not only on the experiment farm but on numerous other farms on the project. Some very definite results have been ob- tained and are being i^repared for publication. It has been found that sodium carbonate (" black alkali "), sodium bicarbonate, and sodium chlorid are the alkali salts chiefly affecting the growth of plants. These salts are found in widely varying quan- tities on different parts of the project, and the proportions in which they are associated are very irregular. Table V states, in percentages, the quantities of sodium carbonate, sodium bicarbonate, sodium chlorid, and other salts found in field S on March 16, 1912, where alfalfa was doing well and also where its growth was not good. This table gives a general idea of the nature and quantities of the alkali salts found on much of the land on the project, but, as before stated, there are Avide variations in this feature. Table V. — I'crcciilitiic of all.dii sdlts tlcicniiiiicd <. HISTORY. The name '' feterita "' is used for a variety of sorg'hum first obtained in 1901 by the Office of Foreign Seed and Phint Introduction as " Feterite" from B. Nathan & Co., Alexandria, Egypt. Only a small quantity of seed was secured, and this was distributed under Seed and Plant Introduction No. 6691 to three persons in Arizona and Kansas, but no records of the results obtained are available. It is certain, however, that the variety did not become established at that time. The second importation, assigned S. P. I. No. 19517, was received November, 1906, from Mr. V. F. Naggiar, of Alexandria, Egypt, who obtained the' seed from Sudan. In 1908 an additional supply, S. P. I. No. 22328, was secured through Mr. R. Hewison from Khartum, Sudan, in which region it is commonly grown under the name " feterita." As it is a member of the durra group of sorghums, the name " Sudan durra '■ has also been applied to it. The value of feterita as compared with the other sorghums of the Sudan is indi- cated by the following quotation from Schweinf urth : - Both varieties of the common sorghum, which here aboiuid in all their minor differences of colour, shape, and size of grains, yield well-nigh a dozen different descriptions for the market at Khartoom. The standard of value is fixed by the Fatareetah, a pure white thin-skinned grain, which also is grown by the negroes in the Seriba. FitzGerald ^ also speaks of feterita as the most valuable of the varieties grown by the natives of Sudan. It has been tested for six years at the Chillicothe Forage Field Station and for five years at the Amarillo Cereal Field Station, both in northwestern Texas. It 1 Issued Apr. 10, 1913. Much of the material for this publication was collected at Chillicothe, Tex., by Mr. G. E. Thompson, formerly connected with the Office of Forage-Crop Investigations. The publi- cation as now presented was completed by the writers after Mr. Thompson's resignation from the Department of Agriculture. - Schweinfurth, (ieorg. The Heart of .Vfrica. Tr. by Ellen E. Prewer. New York, 1874, V. 1, p. 245-24G. 3 FitzGerald, W. W. A. Report on the Improvement and Possible Development of the Cultivable Products of the Sudan, Cairo, 1001!, p. 2."). 25 [Cir. 122] 26 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. lias also hcon leslcd during- shortci- pei'iotl.s at otlu'i' lii'Ul stations in the sonthern Plains area. For the last three years there has been an increasing acreage grown by farmers in that section. The account of the crop given here is based on the results of these station tests and upon the experiences of farmers. DESCRIPTION. Feterita is an early-maturing, drought-escaping sorghum of con- siderable promise both for grain and forage. It has rather slender stems, varying in height from 4 to 7 feet with locality and season. They are somewhat juicy and very slightly sweet before ripening. Suckers are produced freely when moisture is sufficient and are usually taller and later than the main stalk. The plant also branches freely under favorable conditions, and the branch heads are late in ripening. These habits result in considerable unevenness in ma- turit3^ Feterita resembles milo in habit, except that the heads are uni- formly erect and the seeds are larger, softer, and chalk white or slightly bluish white in color. It differs from the ordinary wdiite durra (Jerusalem corn) in having naturally erect heads, black glumes, and plumper seeds. The seed shatters more than milo but less than common white durra if the crop is allowed -to stand in the field until overripe. It is as early or a little earlier than Dwarf milo and two to three weeks earlier than standard Blackhull kafir. In drought resistance it compares favorably with any sorghum yet introduced. In general, the yields have been equal to those of the other gi^ain sorghums and in some cases better, for reasons which are fully discussed later. PROBABLE VALUE. From the foregoing description it will be seen that feterita is more like milo than kafir. It resembles milo in earliness, in the size and height of stalk, in the relatively dry pith, the few leaves, the shape of the head, and the large seeds. It differs from milo most sharply in having the heads all naturally erect and in its larger white seeds. It has become quite popular in parts of northern Texas, because in 1911, a year of severe drought, it produced good gi'ain yields when both milo and kafir gave low yields. In that year much of it was planted on land where the corn crop had been destroyed by drought. In the vicinity of Chillicothe, Tex., many farmers are planting in- creased areas to feterita in preference to either milo or kafir. There is no satisfactory evidence that feterita is inherently more drought resistant than other grain sorghums. The Chillicothe re- [Cir. 122] FETERITA, A NEW VARIETY OP SORGHUM. 27 suits above referred to may have been thie to thinner stands, con- cerning which no exact data were taken. It often happens that thin stands of feterita are caused by faihire of the seed to germinate, especially if planted while the ground is cold. Furthermore, the larger seed of feterita would give thinner stands if planted at the same rate as milo or kafir. At Amarillo, where feterita was ffrown under identical conditions as to stand, it showed no greater drought resistance than milo or kafir. Much interest has been aroused also in feterita bv the extravagant advertising it has recently received from persons either not compe- tent to determine its merits in comparison with other grain sorghums or from those who had seed to sell at fancy prices. Experiments so far indicate that its earliness, its rather low water requirements, its satisfactory yields, and the ease with which it may be harvested give it a real place among the sorghums either for grain or combined grain and forage purposes. Xo farmer should discard milo. Dwarf milo, or Dwarf kafir for fetenta, however, until he has determined with certainty that on his farm it will outyield these staple crops when grown under identical conditions. The data at hand are limited, but they do not justify the claim that feterita will outyield Dwarf milo. PLANTING. Like other grain sorghums, feterita should be planted in rows 40 to 44 inches apart, or about the same distance as for maize, or Indian corn. The time of planting will vary greatly, but, in general, it should be two to three weeks later than for Indian corn in the same season and locality. In sections affected by the sorghum midge very early plantings are recommended. It is not Avell to plant too early, while the ground is still cold, since feterita, like other sorghums, is naturally a warm-weather plant. Owing to the soft seed there is liable to be considerable loss through decay before germination unless the soil is warm enough to sprout the seed at once. Difficulty in obtaining a stand has been experienced in some localities, the poor stand jDrobably being due to planting in cold ground. On the Chillicothe field station, in 1912, plantings of March 19 rotted in the ground, while plantings from the same seed on April 4 and 26 and plantings from a separate lot of seed on May 2 germinated perfectly. At Amarillo a planting on May 11 resulted in a poorer stand and lower yield than plantings on May 22 and 27, about the normal time. Feterita may be planted either in a furrow witli the lister or sur- face planted with the ordinary corn planter, according to the local custom with related crops. When planted in cultivated rows in this [Cir. 122] 28 CIRCULAR NO. 122, BUREAU OF PLANT INDUSTRY. manner 3 to 5 pounds of seed to the acre will be required. The seed should always be sown in a good, firm seed bed and covered with 1 to 2^ inches of soil. In only exceptional cases will it pay to sow feterita broadcast for hay. However, if put in with a wheat drill for this purpose 1^ to 2 bushels per acre will give the best results. CULTIVATION. Feterita should be cultivated in much the same manner as corn. If the ground is in good shape the harrow will be found the most economical tool for the first two or three cultivations. A thorough and i-ather deep cultivation should be given when the crop is about 12 inches high. Two later cultivations will Ije necessary to conserve moisture properly, and in case the ground is foul with weeds still more may be required. In these later cultivations care must be taken that the surface feeding roots are not broken. In order to save all possible moisture it will be found Ijest to culti- vate or harrow after each rain as soon as the ground is dry enough. When the crop has attained such a size that a 2-horse cultivator can not be used, a 1-row, 5-tooth cultivator will be found a very efficient tool. This is also used in many cases to stir the surface quickly after a shower when it Avould not be practicable to cultivate with a 2-horse plow. When the cultivation is finished the ground should be very nearly level. It does not pay to "" ridge the rows," as is often done with corn or milo. HARVESTING. For combined use as forage and grain the crop should be cut in the late dough stage. When planted in rows the crop can best be handled with a corn harvester and ])ut in shocks of 20 to 40 bundles each. In sections where there is little danger of rain the shocks may be larger, if desired. If these shocks are allowed to stand for some time before being headed, it will allow more complete maturity of the heads borne on the suckers. \A^ien the crop is intended solely for grain it should be allowed to stand until the earliest, heads are fully mature. l)ut if left until all the heads on suckers are completely ripe considerable seed may be lost through shattering. Some follow the practice of cutting the heads from the binidles at the time of shocking by holding the bundles across a chopping block attached to the side of the wagon box. The heads fall inside the wagon box and can be hauled at once to a storage shed prepara- tory to being thrashed, while the bundles are set up in the field to be used later. If this practice is followed, the grain should be fully mature at the time of cutting. Another method followed with the [(Mr. 122] FETEPJTA, A NEW VARIETY OF SORGHUM. 29 grain sorghums where h\rge acreages are grown is to harvest the heads by means of a gi'ain header so remodeled as to phice the cutter bar at the proper height. At lower elevations, when the season is favorable and the seed has been planted early, the first crop can often be harvested with a binder and later a good crop of heads obtained from second growth. In an ordinary' year the second growth will furnish abundant pas- ture even though a second grain crop is not produced. Where the heads are cut from the standing stalks m the field it is best to pasture the remainder of the crop. Should the crop go down on account of a storm before it is harvested it can be utilized by pasturing with hogs or cattle. Owing to the fact that the head of feterita is quite heavy in proporti(m to the stalk, the crop some- times lodges badly in wet weather if left in the field until overripe. This makes it desirable to harvest as soon as the crop is mature. THRASHING. Feterita can be thrashed Avith an ordinary grain separator by decreasing the speed of the cylinder and removing some of the con- caves. If a foot or more of stem is left attached to the head, fewer grains will be broken in thrashing. YIELDS. The yields of grain obtained during the five years from 1908 to 1912, inclusive, at the Amarillo Cereal Field Station are presented in Table I. For comparison the average yields obtained each year from the various selections of milo and Dwarf milo are also given. In the first four years the yields of feterita are seen to be fairly interme- diate between those of the other two crops, being higher than those of milo and lower than those of Dwarf milo. In 1912, feterita slightly exceeded Dwarf milo in grain yield. These results may be taken as fairly representative of what may be expected of feterita on the high plains in comparison with the milos. The considerably thinner stands of the feterita are indicated by the row-space figures. LCii-. 122] 30 CIKCULAR NO. 122, BUREAU OF PLANT INDUSTRY. Tabm: I. — 77/f i/roiiiiif/ pciUxJ. roir sjkicc. diid f/rain uidil.s (ihtdhicd from frfvritd at the ccrciil pchl sliilioii, ,\imirillo. Tr.r.. I'.XIS In I'.ll.i, ini-Jiisive, conti)(ircd icith uccnuje siinUur datd for iiiilo and Dicd)f milo. Yvar and variety. Numlier of plats. (! rowing period. Row articles can be made from the long staples than from short. The farmer should receive his share of the advantage, even if it were as easy to maintain the production of long stajiles as of short staples. But in reality this is far from being the case. Long-staple cotton requires not only more favorable natural conditions but greater intelligence and skill on the part of the farmer. If the ability needed in tliis branch of agriculture is not properly remunerated it will not be enlisted. If a new long-staple industry is established, it must be on a very different basis from the present short-staple industry. The results for the first season or two with a new variety in a new district are likely to be entirely deceptive. Unless special attention is given to the problem of maintaining stocks of pure seed by continued selection, the crop is sure to decline in quality as well as in quantity. With the usual neglect of selection, crossing of varieties in the field, and admixture of seed at public gins, the uniformity of an improved stock is usually lost before it has been out of the breeder's hands for more than two or three years. Degeneration becomes apparent in long- staple varieties sooner than in short staples, because the require- ment of uniformity is higher. Spinners can not use mixtures of long and short cotton, so that bales representing a mixture of long and short varieties have less value than the same amount of short cotton alone. It is a mistake to suppose that a successful long- staple industry can be maintained by the ordinary methods of cotton farming. As yet there are very few farmers who appreciate the importance of preventing admixture of seed of different varieties at the gin or crossing in the field. While improvements in this direction may be expected, the change is likely to be very gradual and to become effective only in communities that specialize on the jiroduction of a single superior variety of cotton. If the uniformity of superior stocks is preserved, the seed can usually be sold at a higher price, as well as the fiber. Community organization is as necessary for main- taining pure stocks of seed as for marketing the crop to best advan- tage. Any effective encouragement of long-staple production by commercial or manufacturing interests is likely to come about through cooperation with organized communities of farmers who will limit themselves to a single superior variety of cotton. Desultory plant- ings of long staples by scattered individual farmers are likely to produce only a ])recarious su])i)ly, ii-regular alike in quality and quantity. Improved methods have been developed which simj)lify the work of selection and renchn- it more effective, but all such work must have a basis of thorough familiarity with the variety.^ 'Cook, O. F. Cotton selection on the farm by the oharaelers of ttie stalks, leaves, and bolls. U.S. Dejjartment of Agriculture, iUireau of I'lanl Industry, Circular tiG, 1910. l("ir. 11';!] FACTORS AFFECTING PRODUCTION OF LONG-STAPLE COTTON, 7 PREMIUMS FOR LONG-STAPLE COTTON. Notwithstanding their greater utility the new early-maturing long- staple varieties can not be expected to place the long-staple industry on the same basis as the short-staple industry, and this would not be desirable. The greater care that is necessary in maintaining long- staple varieties, as well as in growing, handling, and marketing the croj). must be compensated by a higher price for the jn-oduct or the precautions will not be observed. The more careful and mtelligent planters will not continue the cultivation of long-staple cotton and permanent long-sta})le communities wall not be established. Nevertheless, the difl'erence in prices between long and short staples need not be as large as was necessary formerly, when only the old late-maturing varieties were available. Figures furnished by an asso- ciation of New England manufacturers, covering a period of years, showed that the cost of their supplies of long-staple cotton had been about 60 per cent higher than the prices paid for corresponding grades of short staples. With the earlier and more productive long- staple varieties that are now available, smaller premiums, perhaps not more than 30 per cent on the average, may be sufficient to de- velop and maintain a long-staple industry. No doubt the actual prices will continue to fluctuate with the seasons in accordance with the general relations of supply and demand, but unless a fair average is mauitained the industry will not attract the particii)ation of the best class of farmers. PROSPECTS OF LARGER DEMANDS FOR LONG STAPLES. If the commercial conditions that determine the demand for lon^- staple cotton remam the same in the future as in the past only a limited utilization of the new varieties is to be expected. With a strictly limited demand a condition of overproduction must be soon reached, prices will decline, farmers will be disaj^pointed, and many of them will go back to short-staple varieties. But in reality the demand for long staples is not strictly limited. It is rather the lim- ited and uncertain supply that has restricted the use of long staples. There is every reason to expect a greatly increased consumption if long staples can be supplied more regularly and cheaply, and this is made possible by the new earl3^-maturing varieties. Of course, too rapid expansion of long-staple production may cause a slump in prices, but a cheap and abundant supply would have the indirect advantage of encouraging use and thus tending to establish a larger permanent demond. Increased demands for long staples are to be expected on account of their industrial superiority over short staples. Short-staple cotton is only poorly adapted for many of the purposes for which it is now ICir. 12a] 8 CIRCULAR NO. 123, BUREAU OF PLANT INDUSTRY. used. Improvements in machinery have made it ])Ossible to spin shorter and shorter fibers, but the resulting fabrics are weaker and less durable. From a general economic standpoint a large part of the labor expended in producing inferior cotton and manufacturing inferior fabrics appears to be wasted, to say nothing of the loss imposed on the consuming public. In \dew of the groat amount of attention that the problem of the increased cost of living is now receiving in all quarters it does not seem reasonable to believe that the immense economic waste involved in the production and manufacture of inferior cotton wnR be left out of consideration much longer. Some of the States have laws to protect the public against textile adulterations as well as against adulterations of food, and the various suggestions for pure-clothes laws are rapidly taking more defuiite form. There is even more reason why the public should protect itself against inferior fiber in cotton fabrics than against cotton mixed with wool or silk, where the deception is much more easily detected. \^^lile there are many legitimate uses for short, inferior fiber, there are also many illegitimate uses, that is, uses not to the advantage of the consuming public, however profitable they may be to the manu- facturer and dealer. In the presence of productive, early-maturing long-staple varieties there is no longer any valid agricultural reason for this wasteful substitution of inferior short staples for industrial purposes that would be better served by the use of long staples. But the mechanical and commercial reasons i-emain, and these will con- tinue effective as long as profits can be made from the manufacture and sale of weak, short-lived fabrics to an undiscriminating public. "VVlien the importance of these factors is once appreciated, remedial measures are not likely to be delayed very long. A large measure of protection would be obtained by a simple requirement that the public be informed regarding the length of the fiber used in textile articles. A regulation of this kind woukl be easy of enforcement, for the length of cotton fibers is readily determined with unproved appliances that are now available. Even though no standard of strength were prescribed, the danger of deception on this side would be relatively small, for manufacturers would hardly find it worth while to make a specialty of collecting supplies of weak long-staple fiber, which would also be more difficult to spin and weave into fine, weU-finishod fabrics. With any ap])roximation to e(|uality in the cost of production the long staples can compete readily with the short staples. The short staples have been used because there seemed to be no prospect of secm-ing adequate and regular supplies of long staples and in spite of the well-known inferiority of the short staples. But now the question is to change back toward the use of long stai)les. The I (Mr. 12:5] FACTORS AFFECTING PRODUCTION OF LONG-STAPLE COTTON. 9 difficulties and uncertainties of the present situation lie in the re- ad] ustnaents that must be made in placmg the cotton industry on a more honest and substantial basis, and thus open the way to a better development of our resources of production. The competition that is rapidly springing up in many foreign countries is likely to force these changes in time, if they are not accomplished voluntarily in the interest of American agriculture and of our own consummg public. CONCLUSIONS. The improved varieties of long-staple Upland cotton developed and distributed by the Department of Agriculture introduce a new factor into the problem of production. On account of their early- maturing habits these long-staple varieties often j^eld as much as short-staple varieties, and they are adapted to cultivation in many parts of the cotton belt where the old late-maturing varieties could not be grown to advantage. Too rapid adoption of the new varieties will result, no doubt, in overproduction and a decline in prices of long staples. But, on the other hand, there is reason to expect a large increase in the demand for long staples, which are better adajjted to many industrial pur- poses than the short staples that are now beuig used. The prospect of a larger and more regular supply of long-staple cotton is likely to lead to more extensive use and tend to more stable commercial con- ditions, thus permitting a gradual development of the new long- staple industry. 8S189°— Cir. 123—13 2 [fir. i:M— B) BEHAVIOR OF SEED COTTON IN FARM STORAGE.^ By Charles J. Brand, Physiologist in Charge, and W. A. Sherman, Scientific Assist- ant, Farmers' Cooperative Cotton Handling and Marketing. INTRODUCTION. The (jiiostion of storing or bulking cotton in the seed for a time before gmnuig is one in which considerable interest has developed. Ginners, spinners in the cotton belt, and others have claimed that many desirable results would follow from a general practice of storing seed cotton for a period of two to five weeks before gmning. They have also emphasized the probable benefits of "sweating out" and aging, and called attention to savings that might result, such as reduction in loss of time of men and teams from standing at the gin awaitmg their turn, increased output of lint per gin stand, and many other benefits. Apparently very few definite experiments along this line have been conducted, and, so far as the writers know, none have been described either m the cotton journals or in the publications of agricultural colleges and experiment stations. In order to determine somewhat accurately the behavior of un- ginned cotton picked at various times and under varying conditions, a cooperative arrangement was made with Mr. E. W. Evans, of Bennettsville, S. C, during the picking season of 1912.2 Mr. Evans had slightly reconstructed one of his farm buildmgs for use in agmg or stormg his staple cotton i)rior to ginning. The dimensions of this buildings are 30 by 60 feet, and the ground floor had been divided by complete jnirtitions into four compartments or bins, each 15 by 30 feet. The arrangement of these bins can be seen in figure 1. The well-floored loft above the storage bins was left without jmrti- tions of any kind and the cotton pickings of each day were placed in the loft, spread generally over the floor to an average depth of 18 inches. The purpose of this procedure was to give the various pick- ings an opportunity to dry out to a moderate degree before they were finally bulked and forked through the trapdoor into one of the bins below. The building has a double wall, being sided with ordinary 1 Issued Apr. 26, 1913. 2 All temperature and moisture oliservations were made liy Mr. Chester A. Kotterman, formerly labo- ratory aid in the Office of Farmers' Cooperative Cotton Handling and Marketing, who was stationed at Bennettsville, S. C, during the greater part of the cotton-picking season. ICir. 123] 11 12 CIRCULAR NO. 123, BUREAU OF PLANT INDUSTRY. weatherboardino- and lined with matched kimber. The inner parti- tions were also of matched material, but of one thickness only. A ventilator about 2 feet long by 10 inches in width, covered with fine wire screen, was placed in the outer wall of the building next to the ceiling of each bin, and the communicating doorways between bins were kept open most of the time, thus aft'ording a limited circulation of air above the stored cotton. On September 30, when tlie final arrangements for the work had been completed, bin A contained a fairly large bulk of seed cotton weighing about 10,000 pounds, com- posed of early pickings. On October 1 this cotton was removetl to bin B, and the first lot of cotton to be used throughout the storage season was forked from the loft into bin A below. Arrangements were immediately made for placing this bulk on a definite floor space, burying in it at selected depths and in difi'erent parts 10 electrical ther- mometer bulbs with leads of suitable length for attaching in the aisle outside to a regular balance indicator with galvanometer. This cotton had been })icked on vSeptember 25 and had re- mained m the loft one week. It weighed 5,800 pounds and when thor- ouglily tramped had a density of about 12.3 pounds per cubic foot in the pile. The average depth of the cotton as fin- ally left for the experiment was 3.5 feet. This body of cotton is designated as pile 1 in figure 1 . On October 2, for purposes of com- parison, a freshly picked lot of cot- ton weighing 6,180 pounds was placed in bin A, shown in figure 1 as pile 2, to be used in comparing the behavior of freshly picked cotton without a preliminary loft chymg with cotton that had been allowed a week in which to dry out. The second lot of cotton was left without tramping and made a pile of an average depth of 4.5 feet, having been so placed as to occupy the same nmn- ber of square feet of floor space as pile 1. This pile of cotton had a density per cubic foot of 10.1 pounds and represented the pickings of October 2 which had been left out in the warm sun throughout the day and then placed in the bin in the evening directly after being [Cir. 123] S/n D < \ift ' 3/n C f \5feet 1" 1 Sir? 3 SiW ^ ^ « 3 -^1 Pile 2 « i-t « soft » Fig. 1.— Floor plan of storage house, showing the size and location of bins and the space occupied by the cotton under observation. BEHAVIOR OF SEED COTTON IN FAEM STORAGE. 13 weighed. The appearance of tliese piles of cotton is shown in figure 2. The mass at the left is the untramped cotton. The tramped pile is shown in part at the back. The leads from the thermometer bulbs may be seen here and there on the face of the cotton. TEMPERATURE IN TJNGINNED COTTON STORED UNDER VARYING CONDITIONS. The records of temperature shown in Table I apply to both piles of cotton, the tramped and the untramped. The bulb buried in pile 1 at point 1 compares most directly with the one buried in pile 2 at •^^ .■.■• PI ^- 0>s .i* ^- • J^Bk1]^^^^^^H p""' •■ •^ L , : ^^^^^^^^^^^^^^» .^ - gW^^ nIsh ij^^^f^Q Fig. 2.— Two piles of experimental coltun in liin A. The one at the left has not heen compaeted in any way; the other was thoroughly tramped. 1', 2 with 2', and 3 with -3'. All were buried about midway of the total depth of the respective lots of cotton, well removed from imme- -diate effects of room and outside temperatures and humidities. The conditions that prevailed in the middle of each pile are repre- sented by 2 and 2'. The temperatures and humidities from October 5 to November 17, shown in the table, are the average of three daily observations taken at 7 a. m., 12.30 p. m., and 6 p. m. [Cir. 123] 14 CIRCULAR NO. 123, BUREAU OF PLANT INDUSTRY. Table I. — Average daili/ temperdtnrrs in trdiuped (ind vntmiiiprd need cotton and average daily temperature (°V.) and relative humiditi/ outdoors and in storage bin, October 5 to November 17, 1912. Temperature (° F.) in tramped (1, 2, 3) and iintraraped (1', 2', 3') pile.s. Outside. Inside. Date (1912). First bulb pair. Second bulb pair. Third bulb pair. Tem- pera- ture. Rela- tive hu- midity. Tem- p<'ra- ture. Rela- tive hu- midity. 1 1' 2 2' 3 3' October— S (i 69.3 69.4 70.7 70.8 71.7 71.3 72.7 73 73.9 74.5 74.6 73.8 72.5 70.9 69.9 70.2 70.2 70.1 69.5 69.7 67.6 66 65.5 63.7 62. S 61. 5 61.2 61. 8 62. 8 62.3 {■i\ 58.9 57 58. 3 59. 9 60.7 60.3 .'^9. 6 58. 9 59. 6 .W. 8 60. 3 h9. 8 57.8 83.1 81.5 78.6 72.8 67.6 68.5 72 72.3 73.2 69.2 64.5 62. 8 60.4 60. 6 65.3 67. ti 64.2 66 67.3 58.8 55.5 56. 7 55.6 53.7 55.1 57.1 59.6 63.1 60.4 50.3 46.7 48.8 57. 6 65 61.6 58. 5 53.8 .52. 6 55. 9 60.9 59. 4 53. 1 48. 6 45.9 69 69.8 70.8 71.3 71.9 71.7 72.2 72.6 73.7 74.8 74.6 73.8 72.2 70.9 70.4 70.3 70.3 70.1 70.1 70.2 67.8 66. 7 66 64.7 63. 2 62 61.9 62.8 62. 6 62. N 61.3 59 57.9 .58. 5 59. 9 61. 1 61). 4 59. 6 58. 9 59. 6 59. 8 61). 3 59. 8 57. 8 82.8 81. 3 79. 9 79.3 77.7 76. 8 76.4 75.9 74.7 75.9 74.2 73.3 71.6 70.1 68. 9 (i9 69.2 68.1 68.3 67.9 66. 6 65.5 64. 9 64 62.5 61.9 61.3 61.7 68. 5 68. 8 70.9 71.3 70.6 70.3 71.3 72.3 72.8 73. 5 71.7 70. 5 6.H. 8 66. 4 66. 8 67. 6 67.7 67.6 67.6 67.3 65 63.5 62.8 61.3 60 60 60.3 61.4 80.4 79. 3 77.9 77.9 75. 6 73.6 73.9 74.5 74.1 73.9 71.8 70.7 68. 7 (;7.4 66. 8 67.4 67.1 66.3 67.2 65.8 64.2 63. 2 62. 5 61.4 60.4 60.2 60.6 61.5 62. 1 59. () ,57. 1 55.9 56.1 .58. 7 60 60 5.S. 1 .57. 4 .57. 9 .59 60. 1 .58.1 .56. 7 54.5 76 74 76 69 65 71 74 .75 75 57 (i3 60 57 63 68 m 60 68 65 53 57 57 57 54 57 62 61 68 50 42 47 51 65 68 57 4(i 49 .53 58 62 (iO 42 42 35 . 55 50 44 (« 61 68 74 70 74 82 69 63 71 79 90 71 89 95 66 52 45 61 50 53 61 69 68 90 55 62 53 56 85 90 61 t)5 61 58 61 74 68 47 62 72 75 76 74 69 66 71 75 74 75 57 63 60 57 63 6S 66 61 69 65 53 55 57 58 53 56 ()3 68 .53 43 46 52 65 69 59 49 51 53 59 64 63 4(i 43 39 58 44 58 S 63 9 53 10 72 11 74 12 74 13 78 14 88 15 69 16 63 17 66 IS. 74 19 90 2f) 71 21 . . 89 22 90 2:3 70 24 63 25 65 2f) 61 27 46 2S 58 29 55 30 69 31 69 November — 1 85 2 t)2. 1 , 62. 3 58 3 61.3 58, ,S 57.6 57.3 58.8 59 59.1 58.4 57.4 59. 9 58.4 59 .57. 6 .57.3 55. 5 59. 8 57 56.1 56.6 58.5 (iO. 6 59.4 58. 9 58. 5 58. 3 59. 4 60. 4 60.5 .57. 6 54. 9 55 4 . . 58 5 63 () 85 81 S 62 y 10 . . 61 66 11 12 58 62 13 74 14 64 1.") K, l(i 17 63 07 The fluctuations in tomporaturo in both tramped and untranipod cotton were so slight that it is evident that practically no heating took place. Bulbs 1, 2, and 3 were in tramped cotton, pile 1. Bulb 1 reache McLachhvn and Mr. E. W. Hudson, both of the Bureau of Phxnt Industry, Mr. Hudson devotin^]: his attention chiefly to the work in the wSalt River Valley. ACREAGE AND YIELDS. Owino: to various causes, a portion of the acreapje planted to Egyptian cotton failed to produce a crop. Among these causes was the lack of irrigation water for some of the fields, while the soil in other fields was too salty, and in one section high ground water, resulting from a flood in the Colorado River, killed the crop in mid- summer. About 480 acres came through the season to harvest. From this acreage 375 bales of about 500 pounds each were finally picked and ginned. Of the total number of bales 262 were ginned in the Salt River Valley, 17 in the vicinity of Bard, Cal., and 96 in the Imperial Valley. These figures do not include 24 bales produced on the experiment farms at Sacaton, Ariz., and Bard, Cal. The yields from different fields varied greatly. Many of the farmers who planted cotton were not familiar with the requirements of the crop, and some fields were given very little attention after planting. The average yield of lint per acre, determined from the total quantity of lint known to have been picked and baled and the total number of acres upon which this cotton was produced, was approximately 400 pounds. The acreage upon which this computation is based included, however, much land on which the crop was very light. A number of fields were measured during the season, and records were kept of the crop harvested. The yields determined from fields in the Salt River Vallev are given in Table I. Table I. Yields of ginnrd cotton from certain measured fields of Egypitan cotton groum ' hi/ fanners in the Salt Rirer Valley in U)12. Sizi' of fu>lil. Yii'ld G ffihor. Total. Average per acre. Acres. I'diinds. Pr,uv(h\ 3.3S 2,527 747 5.32 3,872 729 ,5.35 3,81.S 714 1.42 994 700 4. 75 3,245 (183 l.->.25 lU, 180 007 10.35 5, 937 573 2. S9 1.4M ,500 1.1)2 794 490 13.45 (1,558 4S8 The yields of cotton shown in Table I vary from slightly less than 1 bale per acre to nearly 1^ bales per acre. Wliile these yields were [Cir 12.1] EGYPTIAN COTTON CULTURE IN THE SOUTHWEST. 23 much above the average for the entire acreage devoted to the crop, they are sufficient in number to indicate what may be expected by the better farmers on rich land which had previously produced alfalfa, as was the case with most of the yields reported upon. Two of the yields reported in Table I were those obtained by Pima Indian farmers on the Sacaton Reservation. One of these Indians made a crop of 994 pounds and the other 794 pounds from their fields of about 1^ acres each. Table II. — Yields of ginned cotton from certain measured fields of Egyptian cotton grown hy farmers in the Imperial Valley, Cal., in 1912. Size of field. Yield of fiber. Total. Average per acre. Acres. 11.00 4.76 1.32 5.00 .5.09 Pounds. 6,105 2,537 620 2,000 2,015 Pounds. 555 533 470 400 396 The yields from fields in the Imperial Valley as shown in Table II range much lower than those from the Salt River Valley reported in Table I. The lower yields were due cliiefly to the less careful prepa- ration of the land and handling of the crop by farmers in the Imperial Valley rather than to any essential difference in natural conditions. This is proved by the fact that some of the boys mentioned obtained yields from their half-acre fields approximating Salt River Valley figures. The results of the past summer indicate in a very striking manner the importance of thorough preparation of the land before planting and the need of careful attention to the proper cultivation, irriga- tion, and tliinning of the crop during the early stages of growth in order to secure good yields. It is also clear that while profitable crops of Egyptian cotton may be produced on new land or following grain or a previous crop of cotton, the best results are to be had where cotton follows alfalfa. And, wliile the evidence is not absolutely conclusive, it seems reasonably certain that cotton should not fol- low sorghum or milo, which crops appear to have a depressing effect on the following cotton crop. CHARACTER AND VALUE OF THE CROP. With but few exceptions, the cotton was picked carefully and the seed cotton as delivered at the gins was clean and free from trash. Tliis made it possible to turn out a liigh grade of lint. The bales were well packed to a density of from 15 to 20 pounds per cubic foot, and they were well wrapped with burlap bagging. The average tare, including bagging and ties, was about 14 pounds per bale. (Fig. 1.) [Cir. ll!3] 2-i CIRCULAR NO. 123, BUREAU OF PLANT INDUSTRY. In quality — that is, in len2;t]i and strength of staple — the crop was very uniform and satisfactory. Some of the lint staj)led only If inches, but the bulk of it was fuUy 1-^ inches long and some was shghtly over 1| inches long, Wliilc the entire crop has not yet been sold, the prices for the sales so far reported have been approximate!}' 21 cents per pound, net weight, dehvered on the cars at the shipping point. These prices are said to have been based on the sale of tlie cro}) at 23 cents per pound at New England ])()ints, the inargin of 2 cents being rec[uired to pay the charges for freight, brokerage, and other marketing expenses. TiMi(,am"i»'" Fig. 1.— Bales of Eg\-ptian cotton at Mesa, Ariz., from the crop of 1912. The bales are completely coverecl with a light-weight yet strong grade of lunlap. A small portion of the cotton was of low grad(\ (hie to careless pick- ing, or was of comparatively poor quality, due to bad coiKhtions in the fields. This cotton brouglit a somewhat lower price. The crop from some fields had to be sliipped to assembling })oints to be made up into carload lots, wliicli increased the cost of transportation. However, the bulk of the crop has brouglit tlie producer about 21 cents per pound at his sliipping point, which, consid(M'iiig tlie \Telds previously mentioned, is a satisfactory return. In a few cases the grower sold his crop in the seed; that is, unginned. In tliis conchtion it brought 4f and 5 cents per pound, and at tliis price the purchaser was able to pay the cost of ginning and baling and to sell the fdjer at 21 cents with a safe margin of |)i()fit. M'ii-. i-;;| EGYPTIAN COTTON CULTUEE IN THE SOUTHWEST. 25 From the results of the ginning records mack^ during the season it appears that a Uttle less than 1,800 pounds of seed cotton of the Yuma variety may be expected to give a 500-poimd bale of fiber. There have been marked variations in this respect. The observed range of percentage in fiber from seed cotton has been from 25.2 })er cent to 31.7 per cent, with an average of about 28 per cent for the entire crop. In the Salt River Valley a large proportion of the seed resulting from the 1912 crop has been reserved or sold for planting in 1913. The })rice when sold for this purpose has been from 1| to 3 cents per pound. In the Imperial Valley most of the seed was sold to an oil mill at the rate of $15 per ton. Some of the seed has been used for feed, and when so used was considered as worth about 1 cent per pound. The seed is a factor of appreciable importance in the crop return. Even when sold to the oil miU it brings a sufRcient return to pay more than the cost of ginning and baling the crop. COST OF PRODUCTION. The cost of producing Egyptian cotton, exclusive of harvesting the crop, varied between wide limits. It has been possible in a few cases to ascertain the cost of production; that is, the cost of labor and irrigation water, but not including interest on the land investment. On the larger fields, where teams and machinery could be used to advantage, this cost, exclusive of picking, ginning, and baling, ranged from $11 to $16 per acre. One of the important factors in the cost of production was the preparation of the land. Where it was nec- essar}'' to subdue* a tough sod of alfaKa and Bermuda gi'ass the cost was high. Tliis was particularly the case where the work early in the season was not done thorouglily. The best results in crop yields as well as the greatest economy of labor were secured where the early tillage was thorough and the land was brought into the best possible tilth before the crop was planted. The cost of picking Egyptian cotton was no less variable than the cost of production. On the irrigated land of the Southwest the cot- ton plants gi'ow very large, with many branches. When loaded with a heavy crop the plants bend over and become so entangled that it is difficult to get through the field. (Fig. 2.) Where the acreages were small for each family, no cash outlay for picking was needed. In the Imperial Valley, where labor was scarce and there was a lively demand for pickers in adjacent fields of short-staple cotton, it was sometimes found necessary to pay from 3 to 3 J cents per pound for picking. In the Salt River Valley, on the other hand, the labor supply was adequate and the bulk of the crop was picked for 2 cents per pound. These prices, of course, refer to the seed cotton. It was found that good pickers averaged about 100 pounds of seed cotton per day where the crop was good. In several cases good [Cir. 123] 26 CIRCULAR NO. 128, BUREAU OF PI.AXT INDUSTRY. pickers brought in from 125 to loO ]-)()iinds per day when conditions were es.pecially favorable. Pickers may be expected to average from 70 to 100 pounds per day throughout the season, and when working at this rate the seed cotton should be delivered very clean and free from trash. On the basis of these results it is estimated that one picker will be required for each two or three acres of cotton. The picking season in 1912 began about the middle of September and closed early in February, 1913, though for the best results it should have closed a month earlier, as the late-picked cotton was generally of poor quality. Fig. 2.— a field of Egyptian cotton in Arizona, showing the large size of the plants, which makes picking more expensive than is the case with Upland col ton. The ginning and baling of Egyptian cotton is more expensive than the similar operations for Upland cotton. From the results of last year's experiment it is not possible to estimate with accuracy just what the cost will be when a larger crop is available. The arrange- ments for gmning the first crop were made more with a view to economy in purchase than in operation, and much of the work was done at a disadvantage. From the data obtahied it appears that the ginning and baling cost last year from $6 to $10 per bale. The roller gin used for the Egyptian cotton was the same as the one gen- erally used in the Sea Island district of South Carolina and Georgia. [Cir. 123] EGYPTIAlSr COTTON CULTURE IN THE SOUTHWEST. 27 The turnout averaged about 1 bale per day from each gm, though under favorable conditions it was possible to gin 1^ bales with each machine. The cost of producing an acre of Egyptian cotton, estimating a yield of 1,800 pounds of seed cotton per acre, may be summarized as follows: Seed, tillage, and UTigation, $15; picking, $36; giiniing and balmg, $10; making a total cost of $61 per acre, exclusive of interest on land investment. It should be kept in mmd that these figures are merely approximations. The actual costs will be found to vary between wide limits, both above and below these figures. CONCLUSIONS DRAWN FROM THE SEASON'S WORK. The results of the season's work on P]gyptian cotton in the South- west appear to wairant a material increase in the acreage devoted to that crop. It would appear that farmers, particularly in the Salt River Valley, will be justified in a further trial of the crop and on a much larger scale. The prices paid for the crop were comparable with those paid for imported Egy})tian cotton during the same period of sale. It has been demonstrated by repeated experiments that Egyptian cotton of excellent uniformity and good length and strength of staple can be produced on the irrigated lands of southwestern Arizona and southeastern California. The protluction of Egyptian cotton in larger quantities should result in attracting the attention of users of that staple to this new producing region, and consequently lead to a more advantageous marketing of the crop. The crop is one which fits admirably into the best rotation system for these southwestern irrigated lands. When cotton is alternated with alfalfa the results are beneficial to both cro]:)s and to the pro- ductivity of the soil. Alfalfa is and ])robably will remain the most important single crop of these irrigated lands, but the maximum production of this crop is to be expected only where it is grown in rotation with some cultivated crop, such as cotton, which permits the killing out of the Bermuda grass, Johnson grass, and other weeds which invade the alfalfa fields and after a time greatly reduce their productivity. For the best results in maintaining a liigh quality of Egyptian cotton on these iri'igated lands, it will be necessary to maintain a supply of pure and carefully selected seed. The variety which has been used by farmers during the past season is the result of several years of selecting and testing. The experiments wliich resultetl in the production of this variety are being continued, and other and better varieties may reasonably be expected in the future. Meanwhile, it is of the utmost importance that the seed used for planting by the farmers shall be uncontaminated by cross-pollination [Cir. 123] 28 CIRCULAR NO. 123, BUREAU OF PLANT INDUSTRY. with othor varieties or typos of cotton. In order to insure this, it is very desirable that in each community only one variety or class of cotton be grown. Tliis is important, not only from the standpoint of keeping the seed supply pure, but also in order to avoid confusion in the minds of cotton buyers with reference to the class of cotton pro- duced in each section. It is possible, for instance, for the farmers of the Salt River Valley, by cooperating in maintaining the purity and quality of their seed supply and in carefully handling and harvesting their crops, to develop a reputation for their cotton wliich will be a commercial asset of great value. Such a result would be impossible if several different kinds of cotton were grown in the same region. In the Imperial Valley the size of the land holdmgs and the labor supply are such that it appears desirable for the farmers of that com- munity to specialize on a big-boiled and hence easily picked variety of long-staple Upland cotton, such as Durango; rather than on Egyptian. In the Salt River Valley, however, conditions appear to be more favorable for the development of a cotton industry on the basis of the Egyptian type. There the land holdings are gen- erally smaller, the labor supply more abundant, and it has been dem- onstrated that fiber of very high quality and value can be produced. There remain several important steps to be taken by the cotton growers of the Salt River Valley before the cotton industry can be said to be well established. Larger and more efficient ginning plants must be established, and satisfactory arrangements must be made for the disposal of the seed, either by erecting an oil mill in the valley or by marketing the seed at some outside point. The cotton growers should definitely associate themselves for the purpose of establish- ing and maintammg market grades of cotton which shall be uniform throughout the season and shall fluctuate as little as possible from one season to another. They should also make arrangements for storing the cotton in warehouses after it has been gmned and baled, under circumstances which Avill make it possible to borrow money upon it until it can be marketed to advantage. It is very undesirable to be compelled to sell cotton as soon as the crop is ginned in order to raise money to pay for the cost of production. It is also extremely unportant that every precaution be taken to avoid the mtroduction of the boll weevil and other noxious insects into the cotton-producing sections. It is by no means certain that the boll weevil will thrive under irrigated conditions m the Southwest, but the risk is too great to excuse negligence in this matter. Fmally, it is of the utmost importance to the future of the Egyptian cotton industry m the Southwest that the farmers take up very seriously the question of maintaming an adequate supply of pure seed of liigh- yielding and uniform varieties. [Cir. 123J U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— Circular No. 124. WILLIAM A. TAYLOR, Chief of Bureau. LIBRARK NEW YOjyc BOTANICAL QAKbEN MISCELLANEOUS PAPERS. Agriculture on the Yuma Reclamation Project Effects of Cross-Pollination on the Size of Seed in Maize Experiments on the Decay of Florida Oranges The Wild Prototype of the Cowpea .... CARL S. SCOFIELD f G. N. COLLINS < and [ J. H. KEMPTON J. G. GROSSENBACHER . CHARLES V. PIPER Issued May 3, IQ13. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1913, BUREAU OF PLANT INDUSTRY. Chief of Bureau , William A. Taylor. Assistant Chief of Bureau, L. C. Corbett. Editor, J. E. Rockwell. Chief Clerk, James E. Jones. [Cir. 1J4] 2 ICir. 124— A] AGRICULTURE ON THE YUMA RECLAMATION PROJECT/ By Carl S. Scofield. Agriculturist in Charge of Western Irrigation Agriculture. THE SOIL AND TOPOGRAPHY. The agricultural area of the Yuma Reclamation Project includes about 130,000 acres lying on both sides of the Colorado River in Arizona and California immediately north of the boundary between the United States and Mexico. Of this about 90,000 acres are valley land, formerly subject to overflow by the river, and 40,000 acres are on a bench or mesa directly south of Yuma. This mesa land is about 100 feet above the river and is irrigable by pumping. The valley soil is composed of silt deposited by the river during the annual floods, and much of it is of very recent deposition. The character of the soil with respect to fineness is extremely variable, depending upon the velocity of the water where it was laid down. The irrigable area lies in a narrow strip along the river between low blufl"s of gravel which mark the edge of the mesa. Since 1906 the valley lands have been protected from overflows by levees on either side of the river. Prior to that time agricultural operations on the valley land had been precarious, because of the annual floods. The subsoil underlying the valley is easily pervious to water, and, since it all lies close to the river channel, the ground water is not far below the surface. The height of the underground water fluctuates with the height of the water in the river. In topography the valley lands are nearly level, ^\^th a gentle slope to the south of about 1 foot to the mile. However, there are numerous old sloughs and stream channels left by the river as well as hummocks and sand hills formed from drifting soil. These must be leveled before the land can be irrigated to advantage. This leveling is often expensive, sometimes costing as much as $80 to SI 00 per acre, though the average cost is probably not over $30 per acre. In its natural condition the valley land was covered with a heavy growth of timber and underbrush. The timber consisted chiefly of cotton- wood, willow, and mesquite. The underbrush was arrow weed and a large saltbush. The clearing of the land of timber and under- brash has been expensive, costing from $5 to $25 per acre. 1 Issued May 3, 1913. h'ir. li-'4] 3 4 CIECULAR NO. 124, BUREAU OF PLANT INDUSTRY. The mesa soil is a mixture of fine sand and gravel. It takes Water readily at first, but after a few years of irrigation with the muddy Colorado River water it becomes less pei-vious and tends to pack and bake on drying. Tliere appears to be much less variation in tlie mesa soil tlian in that of tlie valley. Except for a few hummocks, the mesa is smooth, witli a gentle slope to the south and west. It Ues at the east of the valley and is separated froni it by a steep slope. The native vegetation on the mesa is very sparse, being limited to a few desert shiTibs and some inconspicuous annual plants. The cost of preparing mesa land for irrigation will be comparatively shght. About 10 per cent of the valley land on the Yuma Project contains a sufficiently high percentage of alkah salts to render the growing of ordinary farm crops precarious. These salts are composed chiefly of chlorids and sulphates, with very little black alkali. It has been found necessary to provide a drainage system to keep down the water table to a depth from which the capillary action of the soil will not carry the water to the surface. Such a system is now under con- struction. THE CLEVEATE. The climate is warm throughout the year, the temperature for a period of 35 years ranging from a minimum of 16° to a maximum of 120° F. Field operations can be carried on throughout the j'ear, and, if it is desired to do so, the land may be used by growing crops prac- tically all the time. The rainfaU is fight, averaging about 3 inches a year, and is not a factor in crop production. There is marked difference between the temperature conditions on the mesa and in the valley during the winter nights, when frosts occur. The mesa is almost free from frost, even in the coldest weather, while freezing temperatures are of frequent occurrence in the valley during the \vinter months. Such comparative records as are available show that the mean minimum winter temperature is about 9 degrees lower in the valley than on the mesa. AGRICULTURAL DEVELOPMENT. Th(^ valley was fu'st occupied by iVmerican farmers in 1890. The early agriculture was confined to the production of grain and alfalfa. For many years the water supply was precarious and the danger from overflow was such as to retard the permanent improvement of the land. Since the construction of the protecting levees by the Recla- mation Service and with the installation of irrigation works, which insure an adequate supply of water, there has been more diversifica- tion of crops, and much ])ermanent improvement has taken place. Notwithstanding many eiforts to estabfish other agricultural in- dustries, the production of aKalfa has remained by far the most im- [Cir. 124] AGRICULTURE ON THE YUMA RECLAMATION PROJECT. 5 portant industry of the region. Alfalfa fields that are properly han- dled yield annually six or seven cuttings of hay that average 1 to 1| tons per cuttmg per acre. Except during a few brief periods the price of alfalfa hay has been high. The local demand for hay re- quired for the stock used in the construction of the irrigation works has been relatively large up to the present time, but as the construc- tion work is nearing completion the local demand for hay is Ukely to become less active. In addition to the production of alfalfa hay, there has also been an increasingly large production of alfalfa seed. This crop has proved both reliable and profitable. The average yield from 2,824 acres in 1912 was 288 pounds per acre, with a few fields yielding as high as 900 pounds per acre. During the season an alfalfa field produces not only a crop of seed, but four cuttings of hay as well, one cutthig being made before the seed crop and the others thereafter. The alfalfa seed has been clean and of good quality, and it has been marketed to advantage through a local farmers' association. INDUSTRIES IN THE EXPERIMENTAL STAGE. Several crops have been grown experimentally on the project on a scale large enough to merit consideration. These crops include can- taloupes, watermelons, onions, and sweet potatoes. Of the several causes that have contributed to the failure of these crops industrially the two most important ones are high transportation charges and in- experience in marketing. This appUes particularly to cantaloupes and watermelons. In the case of onions the high-priced early market has been occupied by onions from southern Texas, where the crop is two or three weeks earlier than in Yuma Valley, and it has been found possible to ship carload lots of Texas onions through Yuma to Pacific coast markets before the Yuma crop was ready to harvest. Many other crops have been tried on the project, including grain crops, cotton, small fruits, and vegetables. Wheat, barley, and rye are planted m the fall and are harvested early the followmg summer ill time to permit the production of a crop of corn, milo, or kafu' on the same land. Barley is also extensively planted for hay. Alfalfa fields that have been run out by Bermuda grass are frequently broken up in the fall and planted to barley or wheat. Beans have been produced on a field basis, but with few exceptions have not proved a very remunerative crop. The citrus fruits, mcluding oranges, grape- fruit, and lemons, have been grown successfully on the high mesa, while pears, peaches, apricots, plums, figs, dates, almonds, and straw- berries have been grown on a sufficiently large scale to prove the adaptability of the valley lands to their culture. Practically all the common vegetables have been grown with success when put in at the [Cir. 124] 6 CIRCULAR NO. 124, BUREAU OP PLANT INDUSTRY. right season and properly cared for. For mstance, potatoes, cabbage, lettuce, radishes, and carrots are i)lanted during the autunni months, while such vegetables as tomatoes, eggplant, cucumbers, and melons are planted after danger from spring frosts is over. THE KIND OF CROPS NEEDED. Present conditions mdicate that the Yuma Project must develop along the line of producing crop products which are sufficiently high priced to bear the cost of long shipment to consuming markets. It is not probable that the local consumption of crop products will ever be a factor of large importance. The natural conditions of climate, soil, and water supply are such as to favor a very large production per acre. The imi)ortant thing now is to determine which of the many possible crops promise to give the largest returns and to select those which provide for the most suitable utilization of the farm labor and at the same time provide for suitable crop rotation. It is not to be expected that unless a large development of five- stock industries takes place there will long continue a satisfactory market for alfalfa hay. The alfalfa seed crop, however, should re- main a profitable one unless some serious disease or insect pest ap- pears. But alfaha production, whether for hay or seed, is more profitable when the land is occasionally cleaned up by the produc- tion of some cuhivated crop. This is particularly true where, as at Yuma, the alfalfa fields are quickly invaded by Bermuda grass and high production of hay can be expected only during the first few years. Wliile it is true that a fair seed crop may be obtained from alfalfa fields which are badly infested with Bermuda grass, it is inadvisable to allow such fields to remain, because of the increased cost of subjugating the Bermuda soil and fitting the land again for other crops. CROPS SUITABLE TO THE REGION. Of crops suitable for rotation with alfalfa there might be mentioned Indian corn, milo, beans, wheat, barley, and cotton. While such crops as corn, milo, beans, wheat, and barley are all relatively low priced, the climate is such that the small grains can be grown during the winter months and followed by corn, milo, or beans, firing two crops from the same land in one year. Cotton, on the other hand, occupies the land during practically the entire season, but if the proper variety is used the net returns should be larger than for the grain crops mentioned. Cotton is particularly well adapted to rotation with alfalfa because of the high production possible on alfalfa sod. Cotton is especially useful in subjugating land infested wath Ber- muda grass and Johnson grass. It requires comparatively fittle [Cir. 124] agriculturp: on the yuma reclamation project. 7 irrigation but much cultivation din-ing the early summer, and later the plants shade the ground so effectively that Bermuda grass may be completely eradicated by a single crop of cotton. Cotton possesses the additional advantage of relatively high value in proportion to transportation cost, besides finding a ready cash market at all times. It is not quickly perishable and may thus be held without risk in case market conditions at harvest are not favorable. Wliile alfalfa in rotation with cotton may serve as the basis of a profitable agriculture, particularly if accompanied by one or more of the possible animal industries, it is to be expected that several of the more intensive plant industries, such as orchard fruits, will be developed. From present indications it seems probable that certain varieties of pears, such as Bartletts (both early and late). Winter Nelis, and Seckel will do well. The date palm thrives well on the Yuma Project, and as soon as an adequate supply of offshoots of good varieties is available this crop should become a very profitable one. Meanwhile it is possible to purchase seeds in large quantity, and a good proportion of the seedling date palms are likely to yield fruit suitable for home consumption and home market. Moreover, the chance of obtaining valuable new varieties among the seedlings is not to be overlooked. Another orchard fruit of much promise in the Yuma Valley is the fig. Tliere are two important classes of figs, both of which grow well at Yuma — the Smyrna and the common fig. The latter class includes such varieties as the Brown Turkey, White Adriatic, and Black Mission, which yield fruit without cross-pollination. With proper orchard handling two crops of fruit may be produced every year. The fresh fruits of these varieties find a ready and increasingly large sale, and they may also be dried or preserved before selling. The Smyrna fig differs from the common fig in requiring cross- pollination. The fresh fruits are marketed to some extent, but the Smyrna-fig industry is based on the dried fruit. While it remains to be demonstrated whether or not the Yuma Valley is adapted to the production of Smjrrna figs, it seems worth a thorough trial. Figs are well suited to the orchardist operating on a small scale, because the packing operations, whether for the fresh or dried fruit, are relatively simple and inexpensive. The product is also liigh priced, and consequently the transportation charge is proportionately small. CONCLUSIONS. The fullest agricultural prosperity of the Yuma Reclamation Project requhes the utilization of crops which can be combined in suital)le rotation systems and which yield products that can be shipped long distances to market. iCir. 1241 S CIKCULAR NO. 124, BUREAU OF PLANT INDUSTRY. Alfalfa and live stock are likely to be the principal features of this agriculture, at least at first. The orchard fruit crops, such as dates, figs, and pears, may gradually become miportant. AKalfa seed production has been profitable, and it is probable that with more experience in producing and marketing it may be made still more so. Egyptian cotton promises to find a prominent place among the annual crops and is particularly suitable for rotation with alfalfa. A number of other crops are worthy ot trial. [Cir. 124] [Cir. 124— B] EFFECTS OF CROSS-POLLIXATION ON THE SIZE OF SEED IN MAIZE/ By G. N. Collins, Botanist, and 3. H. Kemptox. Assistant, Office of Crop Acclimati- zation and Adaptation Investigations. INTRODUCTION. There is a popular belief that the immediate effect of planting two varieties of maize in close proximity is to increase the yield of both varieties. Failure to appreciate some of the pecuhar characteristics of the maize plant accounts for the delay in bringmg this question to the test. With other crops the changes wliich follow crossmg become apparent only when the hybrid embryos develop into plants m the next generation after the crossing takes place. In maize, however, foreign pollen often has an immediate effect in altermg the color and texture of the seeds, a phenomenon that has received a special name, xenia. It seemed, therefore, not unreasonable to believe that the size of the seed as well as the color or texture might be affected by crossing. In comiection with a study of hybrids between United States varie- ties and a new type of maize from Chma it was observed that one of the effects of crossing the two types was an increase in the size of the seed in the same year the crossing was done.^ Similar results in crosses of tliis Chinese variety have been recorded by Roberts. ^ Additional data on this subject were obtamed in connection with an experiment planned to test the possibihty of selective pollmation in maize. Pollen taken from two plants belongmg to different varie- ties was mixed and applied to the stigmas or silks of one of the varie- ties, thus producmg pure and hybrid seed on the same ear. The phenomenon of xenia made it possible to so select the varieties with respect to color of the seed that it was possible to distinguish between the hybrid seeds and those resulting from the pollen of the same varietv. 1 Issued May 3, 1913. 2 CoUins, G. N. A new type of Indian com from Thina. V . S. Department of Agriculture, Bureau of Plant Industry, Bulletin 161, p. IS, 1909. Roberts, H. F. First generation hybrids of American X Chinese com. American Breeders' Asso- ciation, Annual Report, v. 7/8, p. 374, 1912. 89366°— Cir. 124—13 2 - 9 10 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. The hybrid and i)Uie seeds from each of the ears, when weighed separately, exhibited such striking differences that it is thought advis- able to place the results on record. In every instance the hybrid seed was larger than the pure seed borne on the same ear, the mcrease ranguig from 3 to 21 ])er cent. LIST OF VARIETIES CROSSED. The experiments involved the foUowmg varieties: Missouri Col Pipe, or Collier. — The well-known, large-cobbed vari- ety used in the manufacture of corn-cob pipes. The seeds are white. In our specmiens the average weight per 1,000 seeds is 310 grams. Gracillima. — A variety of pop corn isolated from a variegated vari- ety, the seed of which was originally secured from Germany. The seeds are white. The average weight per 1 ,000 seeds is 67 grams. Variegated. — Similar to the preceding. Isolated from a podded variety. The white seeds of this strain average 141 grams per 1,000 seeds. Hickory King. — A large, thin-seeded, white dent variety. The seeds of the strain used in these experiments averaged 480 grams per 1,000. Mexico Blaclc. — A Mexican variety with slightly dented ''shoe-peg" seeds and an intensely black aleurone. The original seed of this variety consisted of one ear secured from the Valley of Mexico, Seed and Plant Introduction No. 27074. The seeds of this ear averaged 295 grams per 1,000. As growni in our experiments the seeds are smaller, averagmg only 231 grams per 1,000. Algeria. — A variety of pop com with beaked seeds and purple aleurone. The average weight of the seeds is 125 grams per 1,000. DESCRIPTION OF THE EXPERIMENTS. It wUl be seen that the varieties involved comprise four with white seeds and two with colored seeds. The color in these seeds is located in the aleurone cells, the outer layer of the endospenn. Previous experiments have established the fact that when pollen from one of these black-seeded varieties is used to pollmate an ear of the white varieties the resulting seed is colored. In the experiments to be described, the silks of the white-seeded varieties were dusted with a mixture of pollen from the same white variety and pollen of a different variety with colored seeds. The ears that result from an operation of this kind have white kernels that represent pure seed of the variety and colored seeds that are hybrids between the white and colored varieties. In some uistances the ])ollen of the white variety contributed to the mixture was taken [Cir. 124] EFFECTS OF CKOSS-POLLINATION ON SIZE OF SEED IN MAIZE. 11 from the same plant that was behig pollmated, iii which case the result- ing white seeds were not only pure for the variety, but were self- poUmated. The two kinds of seeds were, of course, mixed indiscrimi- nately on the ear and developed under identical conditions. Where consistent differences in the size of the two kinds of seed occur it is therefore safe to assume that they are due to differences mduced by the use of the various kinds of pollen. The number of seeds from a single ear are usually large enough to afford reliable averages and justify confidence in even relatively small differences. Eleven such ears were secured, and the results with respect to the weight of the hybrid and pure seed are given in Table I. Individual plant numbers are given, since in many instances a plant which bears one of the pollmated ears also figures as the source of pollen m another cross. Where more than one ear of a plant was pollinated the second and third ears from the top of the plant are given the individual plant number followed by a figure 2 or 3. To avoid large decimals, the aver- age weight is given in terms of 1,000 seeds instead of in terms of a single seed. It will be seen that in every instance the size of the seed was mate- rially increased by the foreign pollen. The increase ranges from 2.8 to 21.1 per cent. It should also be noted that the rate of increase bears no direct relation to the size of seed in the variety used as the source of pollen. The seeds of the Algeria variety average only half as heavy as those of Mexico Black, yet the increases secured when Algeria is used as the source of pollen are rather greater than when Mexico Black pollen is used. Table I. — Increase in size of seed secured by cross-pollination. Ear-bearing plant. Variety. Pipe. Gracillima . Variegated Hickory King. No. 259 2.39-2 267 1212 1220 I22>-2 122>3 1227 1227-2 1228 219 Source of the mixed pollen. White. Variety. Self. Pipe. Self. . ..do. ..do. Gracillima ...do Self Variegated ...do...... Self Plant No. 267 1212 1212 1228 1227 Colored. Variety. Mexican Black. ...do ...do Algeria Mexican Black. ...do Algeria Mexican Black. ...do ...do ...do Plant No. 643 643 643 1301 669 669 1301 653 653 &53 643 Number of seeds. Pure. 511 541 553 171 99 350 139 402 111 496 140 Hy- brid. 220 248 2J 235 211 94 175 81 24.5 2.5 104 Weight per 1,000 seeds. Pure. Grams. 294 268 371 65 74 66 65 137 140 144 522 Hy- brid. Grams. 337 293 420 79 80 74 74 1.56 148 148 557 Increase with cross- pollination. Per cent. 14.6 ±0.7 9.3 ± 1.0 13. 2 ± .6 21.1 ± .9 7.9 ± .8 12.3 ±1.4 13. 8 ± 2. 1 13.9 ±2.2 5. 7 ± .6 2.8 ± .9 6. 7 ± .2 LCir. 124] 12 CIRCULAE NO. 124, BUREAU OF PLANT INDUSTRY. VOLUME AND SPECIFIC GRAVITY OF SEEDS. The differences in weight observed in these seeds might come about through differences in size or tlirough differences in composition, or a combination of both. To determine this point, vohimetric and specific-gravity determinations wore made of the pure and liyl)rid seed of all tiie ears. The results showed that the observed differ- ences in weight were associated with corresponding differences in volume. No differences in the specific gravity of pure and hybrid seed from the same ear could be detected. SIZE AND COLOR NOT CORRELATED. In order to distinguish between pure-bred and hybrid seed on the same ear only those hybrids with colored aleurone could be utihzed. It might be urged that there was a tendency for colored seeds to be heavier than white and that we liave been measuring differences between colored and white seed rather than dift'erences between pure and hybrid seed. Fortunately, there was material at hand to test this point. In 1911 a cross had been made between Variegated and Mexico Black, two of the varieties used in these experiments. This hybrid was grown and a number of self-pollinated ears secured in 1912. These ears all had both white and colored seeds. A com- parison of the weight of the white and colored seed from each of these ears showed only the ordinary fluctuating differences between the two classes. SELECTIVE POLLINATION AND CROWDING OF SEED. Before accepting the increased size of the hybrid kernels on mixed ears as an indication of the increased yields that may be secured from cross-polhnation, it will be necessary to consider the possibility that the hybrid kernels might develop at the expense of the neigh- boring pure kernels. In that case the average size of the seed would not be increased if all the seeds of the ear were cross-j)oUinated. If the pure kernels are weak or develop more slowly, the hybrid kernels niay gain by being less crowded, or the hybrid kernels may grow more rapidly at first because of more prompt poUination and may then rob their pure-seed neighbors b}^ direct appropriation of a larger share of the available food materials. A fortunate accident in pollination tlu'ows light on both of these questions. An ear of "Maryland White Dent" was polHnated by a plant of the same variety. Seven days later, in making a second a])])lication of pollen, a plant belonging to a red variety with yellow endosperm was accidentally used as the source of pollen, tlie mis- take being noted at the time. [Cir. 124] EFFECTS OP CROSS-POLLINATION ON SIZE OF SEED IN MAIZE. 13 In the resulting ear the two khids of seed were easily distinguished. The pure seeds resulting from the first pollination were ])ure white, while the hybrid seed resulting from the second })ollination were yellow. Unlike the ears where mixed jjollen was used, the two kinds of seed were not indiscriminately distributed. All the white seeds were on the lower portion of the ear; all the colored were on the upper portion. This segregation of the two kinds of seed must have deprived the hybrid seed of any advantage that might be secured by crowding weak neighbors, while the time which elapsed between the two applications of pollen precluded the possibility of the hybrid seed appropriating material in advance of the pure seed. The ear producetl 212 white, or pure, seeds and 161 that were yellow, or hybrid. The average weight of the pure seed was 283 grams per 1,000. The average weight of the hybrid seed was 292.5 grams per 1,000, a difference of 9.5 ±1.06 grams, or 3.4 per cent. The seeds on the lower portion of the ear are usually somewhat larger than those on the upper ])ortion, a fact that should be consid- ered in connection with the observed increase. CONFLICTING RESULTS OF PREVIOUS INVESTIGATIONS. As soon as the phenomenon of xenia came to be recognized, the possibility that size might be among the characters thus affected was perceived by a number of investigators; but, without some method of mixed pollmation similar to that used in our experiments, changes in the size of seed could only be measured by comparing the average size of seed from hybrid ears with the average size of seed in the parent variety. There is such a range of individual variation in maize with regard to the size of seed that hybrid ears would have to be produced in very large numbers to recognize any but very large differences. Since h^d^ritl ears were secured only through artificial pollination, their number was naturally small and it is not surprising that the results were more or less contradictory. Thus Correns^ made a series of crosses between races with different- sized seed and compared the weight of the hybrid and parent varieties. He viewed the results, however, from the standpoint of the inheritance of size as a character and paid little attention to the jjossibUity of this being an uicrease due to crossmg as such. With the limited number of ears which he secured it was hardly to be exj)ected that mcreases of less than 25 per cent could be detected with certainty. In summa- rizing his results on this pomt, he states that while the size of the seed is not essentially changed by cross-poUhiation there is a slight uicrease in weiglit. By the method of m'lxed pollination used in our experi- ments differences averaging as low as 2 per cent are significant. ' Correns, Carl. Bastarde zwischen Maisrassen, Stuttgart, p. 30, 81, 1901. (Bibliotheca Botanica, Heft. 53.) [C'ii-. ]i!4] 14 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. In cases where the ear-bearing phmt i)ro(luced two ears it was hoped that tlie effect of cross-pollination might be measured by pure seeding one of the ears and using foreign ]x)llen on the other. In this way it might.be possible to determine the effects of crossing two varieties witli seed of the same color. But the normal range of varia- tion in the size of seed between the first and second ears of the same plant when both were pure seeded was found to l)e so large as to obscure any differences that might be expected from the use of foreign pollen. ^ NATURE OF THE INCREASE. The Chinese variety in which the phenomenon of increased size was first noticed has very small seeds, suggestmg that the increase that followed cross mg should be looked upon as the immediate inheritance of larger seed of the jiollen parent. The other alternative is to consider the increased size of the seed as connected with the greater vigor so frequently shown m the first generation of a cross. Roberts ' finds difficulty upon morphological grounds in admitting the possibility of this second view. He holds that the growth stimu- lus of the pericarp must be inherited from one of the parents and that additional growth that resulted from a mere increase of vigor would produce an endosperm too large for the pericarp. In the experiments here reported there is no evident relation between the size of seed in the pollen-prod ucmg variety and the amount of increase resulting from cross-pollmation. In fact, the greatest increase was secured by poUmatrng with a small-seeded variety, and there is one instance (plant No. 219) where an mcrease in the size of the seed of the large-seeded Hickory King variety was secured by pollhiating with a variety whose seetJs weighed only half as much as those of the Hickory King. It may be necessary to con- clude that the growth of the })ericarp is stimulated du-ectly by the growth of the endosperm. In experiments thus far reported, how- ever, the necessary increase in the size of the jiericar]) would be com- })aratively slight and to seek any explanation may ])e superfluous. An increase of 22 per cent in the volume of a seed, which is the largest reported, would require an increase in the superficial area of the pericarp of only about 2.8 per cent. If an increase of from 2 to 20 per cent in the weight of the seed can be secured through the stimulation of foreign ])ollen, the fact is of more than scientific interest. Direct e%ddence that important increases can be secured by allowing two varieties of the same color to cross-pollinate has been given Ly Carrier,- who secured yields of 1 Roberts, H. F. 2 Carrier, Lyraaii. Loc. cit. "reventing cross-pollination of corn by means of miLslin screens. Paper read before the American Society of Agronomy, Washington, D. C, Nov. 14, 191'>. [Cir. 124] EFFECTS OF CROSS-POLLINATION ON SIZE OF SEED IN MAIZE. 15 from 5 to 18 bushels more per acre when strains were allowed to cross-pollinate than when cross-])ollmation was prevented. Another idea suggested by the results of these experiments is that increase in the size of the seed in the xenia generation may serve as a means of determinmg in advance the hybrid combmations that will produce vigorous and productive plants the following genera- tion. Whether this proves to be the case or not, the results afford additional reason for the use of first-generation hybrid seed; but even where hybrid seed is not to be used, the planting of two varie- ties in alternate rows may be found to increase the yields sufficiently to warrant the additional trouble. As the increased size is evidently a manifestation of vigor, it may be considered as a factor of adaptation, like the vigor of the first- generation hybrid plants. It would seem especially desirable to take advantage of this method of increasmg the yield in regions which do not produce their own seed corn. CONCLUSIONS. The experiments reported m this paper afford definite evidence that the crossing of two varieties of maize is followed by an increase in the size of seed in the same year that the crossing is done. This increase is not to be confused with the increased yields secured in the year following, when the first-generation hybrid plants are grown. By mixing pollen of a white-seeded and a colored-seeded variety and applying the mixture to the silks of the white-seeded variety pure and hybrid seed are produced on the same ear. By vii'tue of the xenia inheritance of seed color the two kinds of seed produced imder identical conditions can be distinguished and compared. This method makes it possible to measure the immediate increase from cross-pollmation much more reliably and accurately than can be done in any expermients involving the comparison of seed pro- duced on different plants. The results showed the hvbrid seed to be heavier in every instance, the increase ranging from 3 to 21 per cent. [Cir. 124] Cir. 124— C] EXPERIMENTS ON THE DECAY OF FLORIDA ORANGES. By J. G. Grossenbacher, Pnthnlogist, Fruit-Disease Investigations. INTRODUCTION. The oranges that dropped and decayed in many groves in Florida during the past season probably amounted to half of a medium- sized crop, but the trees had been so heavily loaded to begin with that the citrus gi'owers generally had a very profitable year. Drop- ping and some decay first became very noticeable in severely ammo- niated - groves during the latter part of October and gradually increased until the first part of December, 1912, when not only was the ground beneath the trees of ammoniated and melanosed groves covered with moldy fruit, but many trees had large numbers of decaying oranges still attached. The musty, moldy smell of such groves was noticeable at some distance. In a few small sections of the State melanose and ammoniation were practically absent. In these sections the "drops" were compai-atively few, and decay, both on the trees and in transit, seemed no greater than in other years. As might have been suspected, the heavy losses from decay in shipping were chiefly confined to the sections and groves in which the dropping of fruit and decay, both on the ground and on the trees, had been great. During most of December it was not uncom- mon to hear reports of 20 to 60 per cent of decay in transit to market from such sections, and sometimes as high as 80 per cent decayed. The causes of this great loss by dropping and decay are, perhaps, several and may be very complex, but it seems likely that the con- ditions responsible were, at least in part, the same as those causing decay in transit. While the most active agents of decay apparent in the groves during fall and early winter were the blue molds, it is not so evident how these rot organisms gained entrance to the fruits, except ui so far as rinds had spht or had been broken in some other way. In October and again in December, many oranges in very early stages of decay were found still attached to trees, and by careful examination with an ordinary hand lens it was sometimes possible to find one or more very minute clefts in the rind near the 1 Issued May 3, 1913. 2 The words " amnion iai ion " and "melanose" are frequently used in Ihis paper for lirevity in referring to groves having ammoniation or die-back spots and melanose roughenings on the fruit. ' [Cir. 124] 17 18 CIECULAR NO. 124, BUREAU OF PLANT INDUSTRY. center of such a softening area, although in some cases no imper- fections or breaks could be seen. Only rarc^ly could a bright orange be found rotting on a tree, but usually decaying fruit was ammoni- ated or severely melanosed. THE CAUSES OF DECAY. Tliis paper on orange decay as it occurred in Florida during the past season aims only to bring together some of the most ob\nous facts as related to that particular epidemic. The attention is centered chiefly on the weather conditions, cUseases, and other injuries sus- tained by the fruit in the groves and the relation of these factors to decay. As the rot of oranges resulting from improper hantUing is only incUrectly related to decays due to the natural environment it is unnecessary to discuss here the ciuestion of fruit handling. That phase has been fully investigated and some of it has been reported.^ STIGMONOSE AS RELATED TO MOLD INFECTION OF FRUITS. In December, 1912, while looking more particularly for agents that might aid in infecting fruits which have a perfect 'rind, several species of pumpkin bugs or stinkl)ugs were found sticking their long beaks into oranges and apparently sucking the juice from the fruits. It is, of course, uncertain just how effective such punctures are in carrjdng the mold spores wliich may be present on the rind into the fruit, but since the bugs were fairly common and could often be seen making tlu*ee or four punctures in a few minutes it is probable that only comparatively few of their punctures caused infection, in spite of the fact that the mold spores were plentiful throughout the groves. On making hand sections of such orange rinds, especially where very tiny discolored spots could be found in the rag or where one or more pulp cells were found partially empty or cUscolored, it was sometimes possible to get fairly good views of the path of the puncture through the rind by the rod of yellow gumhke substance wliich occupied its place. No incUcation of the puncture was visible on the surface view, even after such a canal or rod had been exposed in section. The sur- face cells seem to close in on the withdrawal of the beak. In a num- ber of such sections of orange rinds the tissue of the inner portion of the outer rind surrounding a bug puncture was found dead and dis- colored, but it was impossible to demonstrate the presence of a fungous mycehum in these inconspicuous early stages. Nevertheless, it is Hkely that tliis stigmonose of the orange was responsible for many mold infections and considerable decay during the height of the period 1 Tenny, L. S., Hosford, G. W., and White, H. M. The decay of Florida oranges while in transit and on the market. U. S. Department AKricultiiro, BiiriMH of I'lunl Industry. Circiil'ir 19, p. 8, 2 fig., 1908. Sec also Ramsey, U. J., l'roceeding.s, Florida State Horticultuial Society, 1912. H'lr. 11^4] EXPERIMENTS ON THE DECAY OF FLORIDA ORANGES. 19 of rot ill December. Infection with mold by means of stinkbiig punctures would, of course, not be evident until the oranges had be- come soft. The manner of infection would also be mysterious, be- cause no evident injury could be found on the rind. Figure 1 is a camera drawing giving a good general idea of these punctures and the early stages of disorganization surrounding them. Possibly some of the disintegration is due simply to a poison se;Creted by the mouth parts of the insects. It so happened that in February, when decay in the groves was very shght as compared to that occurring in Decem- ber, no stinkbugs could be found in the groves. WET WEATHER AND SPLIT- TING INDUCING DECAY. Tlie very wet weather prevailmg during practi- cally the entire summer was doubtless mdirectly respon- sible for most of the rot, not only because it favored the occurrence of both ammoniation and melanose in great abundance, but thereby also resulted in many clefts in the rmd, some large and some small. These rind injuries, in con- nection with the abundance of blue-mold spores, were, perhaps, the greatest im- mediate cause of decay in the groves, as well as of the earlier jiart of the heavy decay m transit. The creasmg and split- tmg of oranges occur com- monly in the fall and winter following a wet season or after a wet period preceded by a drought. A drought checks the growth of the fruit and hardens the rind and, if followed by good grow- ing weather, ruptures may occur. However, when wet weather lasts all summer, amiiKjiiiation and melanose are likely to be abundant ami consequently result in many s])lits. The forces which actually cause the creasing and splitting liave not been determuied, but have been deduced from the conditions of [Cir. 124] Img. 1.— Section of orange rind, showing gummoii.s tnl)e cr rod where i)eak of insect had punctured and disintegration of tissues from within outward. Magnified to aliout 26 diameters. Drawn liy J. Marion Shull. 20 CIRCULAE NO. 124, BUREAU OF PLANT INDUSTRY. weather and growth ])revailing during the seasons when ruptures of the rind are ])revalent; that is, the facts observed liave been ]>ieced together and the gai)s existing in our knowledge liiled ui by assump- tions in such a way as to make a more or less complete story. This theoretical explanation is only provisiona,l and awaits the results of experimental examination before it can be considered a real explana- tion. Many orange rinds sustain ru])tures (hiring or- after a prolonged wet period following a drought which prevailed while the fruit was making the latter part of its growth. The growing weather follow- mg such a drought causes the jnilp to renew its growth and distend the rmd. Smce the tissue of the inner rind, or rag, loses its vitality before that of the outer or oil-bearing portion, its stretching capacity is more limited and, therefore, more often ruptures, while the outer rind becomes smiply more distended. However, when wet weather prevails throughout the summer, as in 1912, the growth of the pulp may be practically continuous, but, owing to the fact that such a wet season is accompanied by the development of ammoniation and melanose, the outer rmd loses its capacity for continued growth and distention and therefore splits. Wlien the outer rind is bright or unmjured, it may be stretched to meet considerable mcrease in the size of the pulp mass; and if ruptures occur under such conditions usually only the inner rind, or rag, is ruptured, resulting m what is known as creased fruit. In either case the rmd of such fruits is found to be very tight or under a very high internal pressure at the time the ruptures occur. If oranges which have an internal growth-and-sap pressure almost great enough to burst the rind were placed in a strong air-tight vessel into which more air is pumped imtil the air pressure inside the vessel is much higher than that outside, the rind on the inclosed oranges would become fairly loose, simply because the air pressure on the outside of the oranges (in the vessel) is now much greater than the growth-and-sap pressure inside the oranges. On the other hand, if, instead of pumping more air into the vessel containing the oranges, air is pumped or drawn out so that the air pressure inside the vessel is much less than the pressure outside, the expansive forces inside the oranges (growth-and-sap pressure) will be so much less restrained by the diminished air pressure surrounding the in- closed fruits that their rinds will burst. In other words, if the pres- sure in the oranges has reached a height wliich is almost sufficient to result in creasing or splitting and then a period of low barometer or low atmospheric pressure comes on, there are almost certain to be many split or creased fruits, while a rise of the barometer or an in- crease in the atmospheric pressure would probably have prevented the [Cir. IL'41 EXPEKIMENTS ON THE DECAY OF FLORIDA ORANGES. 21 ruptures.' The position and tension of the orange rind is a resultant of the interaction between the internal pressure existing in the fruit and the resistance of the rind and atmospheric pressure outside. MELANOSE AND AMMONIATION AS RELATED TO THE OCCURRENCE OF DECAY. As intimated, a rather striking constancy was noticeable in the prevalence of melanose and ammoniation in groves where many fruits were decaying both on the ground and on the trees. It was also noticed that the unusually high percentage of decay of oranges in transit during late November and the month of December was confined almost wholly to those sections of Florida in which the fruits were affected with melanose and ammoniation. In other words, the only sections of the State in which rot was not heavy were the few in which the oranges were neither melanosed nor am- moniated. AVhether it is possible to introduce some factors which would prevent such a wholesale dropping and decay in the groves can not be definitely stated at present, but the question is important enough to merit some very careful study. In a detailed study of the factors in picking and packing fruit which contribute to decay in transit, valuable and suggestive results have been obtained by the Office of Field Investigations in Pomology of the Bureau of Plant Industry. It was demonstrated beyond a doubt that by the older and more careless methods of picking and handling the fruit a considerable amount of decay was induced by injuring the rinds in various ways. But the difference between the decay ordinarily due to such careless handling and that which occurred last year in the oranges from most of the State is very evident and is quite large. A number of carefully run packing houses located in regions where melanose and ammonia- tion were prevalent sustained heavy losses from rot in transit in spite of the additional care taken to avoid the trouble. On the other hand, some packing houses doing business in sections free from melanose and ammoniation experienced but little more than the ordinary amount of rot in transit. In view of such results obtained in the uncommon years, it seems advisable to continue the study of the decay of oranges in transit and to pay special attention to factors in the groves that may be related to the occurrence of such extraor- dinary seasons as the one just past. The unusually high percentage of decay in severely melanosed and ammoniated groves, as well as the enormous loss in transit from such groves when compared with results obtained in groves free from both of these diseases, indicates either that melanose and ammo- niation on oranges predispose them to rot or else that the conditions [Cir. 124] 22 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. favoring the development of molanose and ammoniation arc them- selves the predisi)()sing factors. In quite a number of groves where the splitting, dropi)ing, and decay amounted to perhaps a fifth of the crop, ammoniation was the only kind of rind injury to be found. Once all the details of the cause of ammoniation are known, it may be possible to prevent much of this loss. THE RELATION OF MELANOSE TO STEM-END ROT. In view of the paper recently published by Floyd and Stevens,* m which some suggestive evidence is given to show that melanose is due to the same fungus that causes stem-end rot, it seems higlily probable that by preventmg melanose at least a portion of the decay can be prevented. EXPERIMENTS ON THE PREVENTION OF MELANOSE. Some expermients are now under way on the prevention of melan- ose under grove conditions. Lime-sulphur as well as Bordeaux mix- ture is being used as nearly as possible according to the suggestions given in the above-mentioned paper by Floyd and Stevens. In case the disease is prevalent agam this year it will be possible to get some idea of the feasibility of usmg a fungicide for its control. DOES STELLATE MELANOSE ON LEAVES HARBOR PHOMOPSIS CITRI ? Even in a season like the last, when the leaves, young growth, and fruit of thousands of both pomelo and orange trees are literally cov- ered with melanose roughenmgs, very few fruitmg bodies of the causal fungus were found on the dead twigs of such severely affected trees. That makes one wonder what could have been the source of all the mfectious material required to produce such untold millions of mfec- tions. It appears that ordinarily PJiomopds citri is not to be found in the melanose roughenmgs on the leaves, shoots, or fruits. The mjury resulting in the rough spots is thought to be duo to some enzymotic substance liberated by the spores of the fungus on germma- tion. Tlie melanose spots are usually pinheadlike cushions of cork raised above the general surface of the citrus tissues, but hi some cases a later enlargement of these circular spots occurs, in that three or more short ridges develop as radii about the origmal circular si)ot. This stellate form of melanose was found very abundantly durmg February and March, 191.3, ui one of the smaller orange groves of the Manatee Fruit Co. near Palmetto. Now, smce the fruithig bodies of the Phomopsis are so few on tlie dead twigs of severely affected trees, 1 Floyd, B. F., and Stevens, H. E. Melanose and stem-end rot, Florida Agricultural Experiment Station, Bulletin IH, 16 p., 9 figs., 1912. ICii-. 124] EXPERIMENTS ON THE DECAY OF FLORIDA ORANGES. 23 it was thought possi])le that under certain unknown conditions the fungus may actually be present in the original melanose spots and by further growth mduce a development of stellate spots and fmally produce spores and thus mfect the new growth of leaves, fruits, and shoots. These stellate melanose spots have been studied carefully in thin sections, but fungus mycelium could not always be found in them, although some was often present. The accompanying illus- FiG. 2. — Stellate melanose spots on upper side of orange leaf, showing the older circular spots in the center. Magnified about 7^ diameters. Photographed by Dr. Albert Mann. trations (figs. 2 and 3) give a fair idea of the stellate spots as they appear under an ordmary magnifying glass. EXPERIMENTS TO DETERMINE THE CAUSE OF DECi».Y. Since the general opinion prevailed, both among the growers and shippers, that the rot was chiefly due to Penicillium or mold and because the heavy decay was confined to groves that had melanosed and ammoniated fruit, it seemed worth while to determine what relation could be shown to exist between those rind diseases and decay. A TEST AT ORLANDO. In order to make a preliminary test, bright, ammoniated, and melanosed oranges were carefully picked in early February near Orlando, Fla., and sprayed with blue-mold spores obtained from hly decayed oranges secured from a packing house near the Ml-. 124] 24 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY, Orlando Field Laboiatoiy. After drying, (ive l)riglil, five melanosed, and five ammoniated fruits were wrapped and i)ut into separate cov- ered glass vessels. The vessels were stored on shelves in the labora- tory. Tw^elve days later the following results were found: In the dish containing the melanosed oranges one was about half decayed and two others were slightly softened at the stem end, but no blue mold was evident. In tlie dish containing the ammoniated fruits two were entirely decayed and the blue mold had covered the outer surface of the paper wrappers with its spore masses ; another orange Fig. 3.— Stellate melanose as occasionally seen on t he under side of leaves. Magnified about 7 \ diam- eters. Photoj;raphed by Dr. .\lbert Mann. was about one-fifth decayed at the stem end, but had no blue mold in evidence. The bright fruits were apparently as good as when picked two weeks before. Although these numbers are small, the results are suggestive, because they are so pronounced on fruits which had been carefully picked and handled. The difference between the 60 per cent of decay of the rough-rinded oranges and the entire absence of decay in the bright fruit was so great that it was decided to make further tests. It appeared rather extraordinary that half of the oranges were decaying by an organism distinct from the one used in the water sprayed on the fruit. [i-n: 1-J4] EXPERIMENTS ON THE DECAY OF FLORIDA ORANGES. 25 THE FIRST TEST IN WASHINGTON. It also seemed possible that oranges from a vigorous tree might prove more resistant to decay than those from a tree that had been much weakened by crown-rot (foot-rot), even when the rind of both is in about the same condition. In order to open the way for some future work along that line, melanosed fruits were selected from both normal and foot-rotted trees and carefully marked when pack- ing. Ammoniated fruits were taken from vigorous-looking trees. The fruit was carefully packed and expressed to Washington, D. C, from Orlando, Fla., on February 28. Owing to unforeseen conditions, it had to be held in Washington at room temperature until March 10 before being unpacked to start the experiment. At that time several ammoniated oranges were found decayed, apparently by blue mold, and several melanosed fruits and one bright orange were in the early stages of stem-end rot, as indicated by obtaining PTiomopsis citri from their interior. A number of melanosed and ammoniated fruits were rather soft and flabby. Only the firm ones were used in the expernnent, the details of which are given in Table I. A portion of each type of fruit was sprayed with water by means of an atomizer, and the other fruit was spra3^ed with a suspension of the Penicillium or mold spores obtained from vigorous pure cultures formerly secured from moldy oranges. The mold used seems to be most like Penicil- lium italicum, but at present its identification is uncertain. After drying, the fruits were wrapped with ordinary orange wrap- pers and each kind placed in separate covered glass dishes. The fruit was all held at room temperature for three days and then some was transferred to a refrigerator having a temperature of about 15° C. (60° F.). On March 17, or a week after the beginning of the experi- ment, the oranges were all carefully examined for decay, and cultures vrere made from the interior of many of the fruits which seemed to be affected by stem-end rot, although the decay was not of sufficient extent to be certain. All but two of such doubtful ones were found to contain Phomopsis citri. The fungus obtained from the two other oranges proved to be neither Phomoj)sis nor Penicillium, and not having produced its spores as yet in culture it still remains unknown. The relation of Phomopsis citri to stem-end rot was established b}^ Fawcett/ who also showed that the decay is not preventable by the application of a fungicide. 1 Fawcett, H. S. Stem-end rot of citrus fruits. Florida Agricultural Experiment Station, Bulletin 107, 23, p. 9 figs., 1911. ICii-. 124] 26 CIRCULAE NO. 124, BUREAU OF PLANT INDUSTRY. . "O -J -«■) e*ip*5 •E _: ^ t- C Oi I 10 6 o tjco ^ :DC 5 t* i S C"! »C C 1 CO . .-1 '* -^ T a. M m < a : c3 ; -^ o - tf « ?i w ":> -1*0 J rci 5 A -d lO t^ In -. M <■ <— 1 c^ o 10 IM Oi 3 t~ : ^ ^z;"^ 3 GC •*• M ■H r-.CCC 3 00 1^ c^i c^ 'J^ HM P 1 _C3 t. -*->* fl »o c ? § : , ^"=^ 3 1 c-l w) t^g ' s CO o o I . — 1 1 S j coc > OC 3 s 00 o yo £ C3 o s d 03 0^ 5 J^ ? fi '? X) c -« w ^ 1^0 1 fd .-H r-l C ' >ntj (U CN V « 4^ c- 0) Ut >> ^ O -M c3 w 3 a C3 Xfl u7 c -t-a X P. p. 5 o o -^ "■ v^ ^3 '^ lO C > C33 03 Q S ii'o •0 »o - li 3 a> ^^ rJ c; ttCv > 1 M 000 Ct-iJ a^E ta 3 ^ 5^ 'E t^ C u I-. u « ■3 a* u, ■a' Mfl 1 1 S m oc > OC c3 IS >> 8 a; 5 000 00 •o p; O "t^ £ ft 03o 1 a ■M - "S 0. a H 00 1 00 1 — Gi _^ P -4-^ a ^' a .M CO 22 '3 5 oj « lO ir ) 3 ill 2; 3 cc >o -■' C-1 >-H , e*o Mn , . e«- t- +^ c lO C > IX. u< ooc ID Qi C !>«:; .r > tP c; CC«5 iT '^ P. fc p. — s ^ a) O •3" a) c3 S c 001 1—1 1 -g S 00^ 1— 1 CO +J .g 1 5 +^ A -*-» § 5 3 o -3 a C3 ft 3 '? Q "~" Si Ml 3 O sa e " fc X' ir be 3 — .§ P- 02 CO^ -* coco c ^ oc -4^ 1^3 00 c CO — t- a> lI 0^ s sss s.a s SSS s o o Q ■2 ■3 » o3 s s fH ^ CN -^ 3 00c 00 C3 '? C3 ■M ■3 H -f ^ ■HP 5 -4J ft n w C^l rf ^ 03' CO C>4 « ^ t) •fi M "T s P. 3S? 7^ 3 »r: Tf iO If- 01 "3 - hi :5'~ 3 »c »o u: 10 +^ ■*_» ■♦-» ^ •4^ s '^ » ^H C-l 5 bZ 1 ^ t-i ft O C3 «s s a> 'O ^ ■*j H »i3 6 h'o P C9 c 2« 5S 1 O p 1 , 0-3 1 ~ a a "5 1 •• 1^ on "3 1 - n S P. ■g-s i| C3 3 3 ' g- g^ ■< 1 w 1 H H 1 2 ■3 a 3 ,Q 3 o -3 >. C3 •3 II pq 2 "3 a >. .p 3 o S ■3 a> >> OS tCir. 124] EXPERIMENTS ON THE DECAY OP FLORIDA ORANGES. 27 Several interesting facts are evident from the data recorded in experiment 1 of Table I. Perhaps the most strilving thing here is that fruits treated with mold spores were but little more subject to decay than those not so treated. Another important pomt that may have a wide practical bearing if further experiments confom it is the fact that even though three days were allowed for the incubation of the oranges treated with blue mold before they were removed to a temperature of 15° C. (60° F.), no case of blue-mold decay resulted in the refrigerator in a total of 19 sprayed with spores. Of the total number of fruits decayed, only 28f per cent were rotted by blue molds, while 7 If per cent decayed by stem-end rot. Two of these fruits rotting at the stem end were afterwards shown to contain a very rank-growmg fungus which has not yet s])orulated in culture, while the others contamed Phomopsis citri. It is also worth noting that of the oranges sprayed with Penicillium spores and held at room temperature throughout the experiment, 38^ per cent decayed at the stem end and only 10^ per cent had blue mold. Of the total of 58 oranges treated with mold spores, 24 decayed; of these, 20, or 83 J per cent, were decayed by Phomo])sis, while only 16§ per cent of them were rotted by blue mold. THE SECOND TEST IN WASHINGTON. In order to get an idea of any change that might occur m the per- centage of decay as the season advances and to study further the effect of refrigeration as a preventive of rot, another test was begun on March 28. Mr. Leslie Pierce, who was supervising some spraying work on the prevention of melanose in Florida, picked and sent to Washington, D. C, from Orlando, Fla., melanosed, ammoniated, and bright oranges to be used in the experunent. In this case some of the fruits were sprayed with the spores of the common citrus blue mold, Penicillium italicum, and others with a dark olivaceous species also found on oranges and which appears to be Penicillium olivaceum. But owing to the fact tliat the application of mold spores seems to have had so little effect the two fungi are not distinguished in the records as given in experiment 2 of Table I. The data in this second experiment are again rather surprising, both in regard to the relative numbers of oranges decayed by Phomopsis citri and by mold and in regard to the effects of only moderately low temperature in checking or preventing the decay. Of 30 fruits sprayed with spores and held at room temperature, 14 decayed; 13 of them, or 92f per cent, were decayed by Phomopsis citri, and only 7^ per cent by Penicillium. Only 5 of the total 17 held in a refrigerator had been treated with mold s])ores; 1 of the 5, or 20 per cent, had a very sliglit indication of decay at the stem end. This [Cir. 124] 28 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. would tend to strongtlien the evidence aiTorded l)y the former ex- periment in making it aj^pear wortli while to precool oranges for transportation when they are taken from groves which yield a high percentage of decay. SUMMARY AND CONCLUSIONS. These preliminary experiments are too small and few to permit any but tentative conclusions, yet the results are so clear cut and uniform in regard to some of the significant points that they may be specially noted. In the first place, it should be borne in mind that the condition of the fruit at maturity is a resultant of the environ- ment of the grove during the growing season. Coincident with the uncommonly wet season of 1912 melanose and ammoniation of the fruit were prevalent over most of the citrus region of Florida, and it is very probable that the wet weather was directly related to the occur- rence of such rind injuries. The excessive moisture of both the soil and the air in connection with these injuries resulted in numerous splits and rind distentions, which caused an enormous development of mold in the groves. The presence of large numbers of stinkbugs^ which suck juice fi'om oranges, probably also added to the epidemic by inoculating sound fruits when inserting their beaks through portions of rinds on which the air currents had deposited mold spores- The rather muggy warm weather prevailing in late November and in December doubtless also contributed to the heavy decay of that time, for these experiments indicate that temperatures no lower than 15° C. (60° F.) practically prevent the development of the mold rot. But in addition to the mold-rot epidemic induced by the split and cracked rinds, much fruit dropped and decayed which was not split. The above results make it appear probable that Phomopsis citri was responsible for most of the rot of that type except w^hat was due to bug-puncture infections. This also agrees with the fact that melanose was very abundant in the districts where much fruit de- cayed. However, the scarcity of phomopsis pustules on the dead twigs during late fall and winter argues that this fungus may sporulate elsewhere or that its mycelium persists in some melanose spots on the stem or calyx of the fruit. Perhaps the most important conclusion that these observations force upon us is the importance of growing fruit that is free from melanose and thus not only obviating a reduction in the market value of the fruit but also preventing some splits and most of the stem-end rot. The fact that the refrigerator temperature used in these experiments ])ractically prevented all decay suggests tliat shipping tests should be made with precooled fruit in refrigerator cars and boats. ICir. IL'4] [Cir. 124— DJ THE WILD PROTOTYPE OF THE COWPEA/ By Charles V. Piper, Agrostologist, in Charge of Forage-Crop Investigations. In a recent bulletin the writer has given an extended account of the cultivated forms of Vigna known as cowpeas, catjangs, and asparagus beans, all the data and conclusions being based on cultivated material.^ In that bulletm the three were considered different species, namely, Vigna sinensis, V. catjang, and V. sesquipedalis, but it is pointed out that all can be crossed readily and that a perfect series of inter- grades exists in respect to all characters. In a previous bulletin Mr. W. F. Wight ^ has gone into great detail mto the early history of this same group, reaching the con- clusion that they represent three distinct species and — That both Vigna sinensis and V. catjang originally came from the region including and extending from India to Persia and the southern part of the Trans-Caspian district, and that the Persians called one or both of them by the name ' ' lubia ' ' and applied that name to V. sinensis in northwest India after their conquest of that region. The cultivation of V. sinensis extended to China at a very early date, but the distribution of at least one of the species with the name "lubia " has extended from the region of its origin at the beginning of the Christian Era to Arabia and Asia Minor and had reached some of the Mediterranean countries of Europe at about the same time, but did not become known in Central Europe until the middle of the sixteenth century. One difficulty in accepting Wight's conclusion as to the original habitat of the cowpea is the fact that no wild plant is known in the region indicated that could in any likehhood be the wild prototype. This difficulty is, however, paralleled by a similar one in the case of maize, kidney bean, and other plants cultivated from remote an- tiquity. Still earUer the subject had been discussed by Kornicke,* who reached the conclusion that all the cultivated forms known as cowpeas, catjangs, and asparagus beans represent but one botanical species, Vigna sinensis, whose native habitat he befieved to be central Africa. He based this opinion on specimens collected by Schweinfurth, by Schimper, and by Kotschy, but enters into no botanical discussion regarding the identity of the wild with the 1 Issued May 3, 1913. 2 Piper, C. V. Agricultural varieties of the cowpea and iramediately related species. U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 229, 160 p., 12 pi., 1912. 3 Wight, W. F. The history of the cowpea and its introduction into .Vmerica. U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 102, pt. 6, p. 59, 1907. ^ Komicke [Ueher von Apotheker Winter bei Gerolstein aufgefundene neue und seltenere Pfianzen.] Verhandlungen, Naturhistorischer Verein der Preussischen Rhemlande, Westfalens und des Reg.-Bezirks Osnabriick, Jahrg. 42 (F. 5, Jahrg. 2), Correspondenzblatt 2, p. 136-153, 1885. [Cir. 124] 29 30 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. cultivated plant. Indeed, it is not clear that he examined the actual specimens, as he writes, ''I predict that the wild forms are slender and twining and bear pods which, when ripe, sprirg open." (Translation.) A recent opportunity to study the Yigna material in the Berlin, Kew, and British Museum herbaria and the study of the wild African plant under cultivation has led to the conviction that Kornicke's conclusion regarding the native plant of Africa being the wild form of the cultivated cowpea is correct. The wild species itself is quite variable, and its relationsliip to other supposedly distinct species is not clear. Until these supposed distinct species are also studied under cultivation there must remaui some doubt as to their actual affinity. From herbarium studies the following facts are adduced and conclusions offered: Southward from the Sahara Desert and extending across the con- tinent is a wild plant that differs from most cultivated catjangs only in th€ following characters, viz: The leaflets are minutely sca- brous on the upper surface and the petiolules are usually pubescent ; the small pods are dark colored, scabrous, 7 to 8 centimeters long, and in dehiscence the valves coil tightly. In herbaria this plant is usually labeled either Vigna sinensis or Vigna nilotica, and on one of Schweinfurth's sheets the name Vigna spontanea is written. The numerous collections of tliis plant include specimens from Egypt, Nubia, Kordofan, Abyssinia, German East Africa, Zanzibar, Senegal, Gold Coast, Kamerun, Nigeria, Angola, Rhodesia, Natal, and ^ladagascar. It is sometimes cultivated, as indicated by Schweinfurth's No. 1778 collected between Debenhuch and Ken eh, Egypt, and No. 888 from Khartum. Lederman also found it cul- tivated at Garua, Kamerun, as sho\m by his specimen No. 5149a. Stuhlman's collections in Zanzibar give the native name as "kunde" or "kunde ya muita," and state that the wild bean is not eaten. The name "kunde" is, however, generally appUed to the cultivated cowpea in German East Africa, or, according to Braun,^ the seeds are called "kunde" and the plants "mkunde." In northern Nigeria the wild plant is called '•gayan-gayan" and seeds from specimens collected in Sokoto, Northern Nigeria, by Dr. J. M. Dalziel (No. 318, October, 1910), germinated freely and grew in a greenhouse in Washmgton. The nearly black scabrous pods are about 10 centimeters long. The seeds are 5 millimeters long, buff, marbled with brown, speckled with minute blue dots, and have a few black blotches. Mr. George W. Oliver has already crossed this plant with various varieties of cowpeas. Prof. W. J. Spillman hazards the prediction that the wild ])lant contains in its seed colors all the 1 Braun, K. Bestimmimijstabi'Ik'n fur dw Eingeboreneiikullumi von Ueutsch-Ost-Afriku. Der pflanzer, Jahrg. 7, No. 8, p. 440, 1911. |Cir. 1-4] THE WILD PEOTOTYPE OF THE COWPEA. 31 color factors found in the numerous cultivated varieties, except perhaps the "eye." The hybrids above mentioned have been made partly with the idea of clearing up this point. The numerous cultivated forms of Vigna in Africa are mostly true cowpeas, and about 40 varieties from Egypt, Sudan, Rhodesia, Transvaal, British East Africa, and German East Africa have been grown in comparative trials and are described in Bulletin 229 already cited. In Angola occur cowpeas distinguished by having more or less retrorse pubescence, especially on the stem. These have not been grown in our trials, but excellent botanical material was col- lected by Welwitsch, who considered the plant a distinct species, Vigna macundi. Except for the small amount of pubescence, how- ever, there is nothing to distinguish it from an ordinary cowpea, Welwitsch ^ speaks of it as commonly cultivated and occasionally spontaneous. About Golungo Alto, Angola, the native name is "macundi" (plural) and "hcundi" (singular) both suggestive of "kunde," the native name of the cowpea in German East Africa. The asparagus bean apparently is rare in Africa, only one lot, S. P. I. No. 11091, from Abyssinia, having been secured. The catjang is not infrequent in Africa. In Bulletin 229 four varieties from Abyssinia are described, and Schimper found it abun- dantly cultivated near Humboldt Springs, Djur Land, Anglo-Egyptian Sudan. From the wild cowpea the catjang differs mainly in having the pods pale and smooth instead of dark and scabrous, but all inter- grades occur. A form with pale, scabrous pods was collected by Ehrenberg in cultivation at Gumfuda [Kunfuda], Arabia. In all probability this is the same as the Egyptian plant named Dolichos luhia by Forskal, also described as having scabrous pods. Similar specimens have been collected in the Seychelles (I. Horn, No. 474, in 1874) and in Java (Surokarta, Horsfield). Such specimens also come from Meslira el Zaraf, Sudan, with the mformation: "Eaten by cattle; seeds eaten by natives during the year of drought; native name, 'Lubiet el Gazal.' " Besides the above-mentioned plants, all of which are undoubtedly referable to Vigna sinensis, there occur in the southern half of Africa other forms where relationship to the cowpea is less clear. The commonest of these is the plant known botanically as Vigna triloba Walpers, which differs from the wild cowpea mainly in its leaflets being almost always three lobed. Some of the labels mdicate also that the plant is not a true annual, but from herbarium speci- mens this can not be determmed. Unless there is a distinction in the matter of life period or duration, Vigna triloha can not be kept distmct from the wild form of Vigna sinensis by any character yet 1 Hiem, W. P. Catalogue of the African plants collected by Dr. Friedrich Welwitsch in 1853-61, pt. 1, London, 1896, p. 260. [Cir. 124] 32 CIRCULAR NO. 124, BUREAU OF PLANT INDUSTRY. pointed out. The lobing of the leaves is very variable and commonly occurs in cultivated Vigna i^ineni^is, especially when in poor soil. Vigna triloba occurs mainly south of the P^quator, where it is appar- ently never cultivated. However, specimens collected by Dr. J. M. Dalziel in the Katagum District, Northern Nigeria, May, 1908, wliich seem indistinguishable from Vigna triloba, bear the legend ''Common cultivated bean of the fields." In Angola, forms of Vigna triloba with narrow leaflets occur. Such specimens are represented by Welwitsch's No. 2262 from Loanda and No. 2263 from Pungo Andongo. On the former of these Wel- witsch has recorded "a herb persisting for several years but scarcely ])erennial," and on the latter "a herb enduring apparently for several years." Gossweilcr has more recently made collections in the same region. His No. 4802 from Grongude, apparently identical with Wel- witsch's No. 2262, is stated to be "a perennial many-leaved vine, ^\dth large blue flowers," while liis No. 1534 from Penodo, quite the same as Welwitsch's No. 2263, bears the legend "u perennial cUmber." Differing from this narrow-leaved plant mainly in being pubescent is a form occurring in Natal, Transvaal, etc., which was called ScytO/- lis protracta by E. H. F. Meyer. All recent botanists have consid- ered tliis only a form of Vigna triloba. Another form with still narrower leaflets and lacking the pubes- cence has been named Vigna triloba stenophylla Harvey and Sonder.^ In the light of present evidence it is not clear that Vigna triloba is specifically distinct from the wild Vigna sinensis of Africa. On the other hand, the statements of Welwitsch and of Gossweiler that Vigna triloba persists more than one year indicate that it is different, in this respect at least, from Vigna sinensis. The trilobed leaflets as a mark of distinction can not be depended upon, yet in classifying specimens this is the only evident character. In the herbarium at Kew are specimens of both species', so far as this distinction may be depended upon, collected by Dr. J. M. Dalziel in the Katagum Dis- trict, Northern Nigeria. The one referable to Vigna triloba is said to be ''the common '6ultivated bean of the fields,'* wliile no data regarding the habitat, of the other are recorded. There is scarcely room for doubt, however, from the ani})le herba- rium material, that the ^ifrican plant with blacldsh scabrous pods and scabrous leaflets found wild over a great area and occasionally cultivated is the original wild form of our cultivated cow^ea. This conclusion is further confu-med by a study of comj^arative cultures and the fact that hybrids between the wild plant and the cowpea are readily obtained. Whether Vigna triloba is to be considered a distinct species can hardly be decided in the light of j^resent evidence. ' Harvey, W. H., and Sender, O. W. Flora Capensis, v. 2, Dublin, 1861-2, p. 241. I (Mr. IL'4] o UBRARY NEW YOKJK B<^TANICAL GARDEN Issued May 7, 1913. U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— Circular No. 125. WILLIAM A. TAYLOR, Chief of Bureau. SUDAN GRASS, A NEW DROUGHT-RESISTANT HAY PLANT. BY C. Y. PIPER, Hgrostologist in Charge of Forage-Crop Investigations. 82199° — 13 WASHINGTON : GOVERNMENT PRINTING OFRCE : 1913 BUREAU OF PLANT INDUSTRY. Chief of Bureau, William A. Taylor. Assistant Chief of Bureau, L. C. Corbbtt. Editor, J. E. Rockwell. Chief Clerk, Jambs E. Jones. tar. 125] 2 B. P. I.— 839. SUDAN GRASS, A NEW DROUGHT-RESISTANT HAY PLANT. INTRODUCTION. For several years past, beginning with 190G, the writer and his assistants made a careful stndv of Johnson gi-ass wdtli the view of finding a strain lacking the underground rootstocks which make Johnson grass so objectionable. While variations in this character were found, no single plant was detected which had the rootstocks wholly absent. Coincident with these studies packages of Johnson grass seed were obtained from various foreign sources, in part wdth the assistance of the Office of Foreign Seed and Plant Introduction. Among those received are two other varieties bearing rootstocks like Johnson grass but differing in other characters, and two very distinct varieties that have the rootstocks wdiolly absent. The first of the lat- ter Avas received in 1909 under the name " garawi," through Mr. R. Hewison, Director of Agriculture and Lands of the Sudan Govern- ment at Khartum. After growing this for one season at Chillicothe, Tex., it was inventoried as Seed and Plant Introduction No. 25017. In further correspondence with Mr. Hewison some additional in- formation has been secured. The Sudan botanists were under the impression that garawi is a form of Andropogon Kalepensis^ or John- son grass. According to Mr. Hewison, the following note appears in Broun 's Catalogue of Sudan Flowering Plants : Andropogon haJcpensis Brot. Adder or Adra (wild variety) and Garawi (cultivated), Arab. Tall grass cultivated for fodder. The seeds are eaten in times of scarcity. When wild it grows to a height of 12 feet and is found in damp localities along the river banks or edges of pools. Found in Sennar, White Nile, and Kordofan. Whether the wild plant is the same as the cultivated Mr. Hewison is not sure, and promised specimens have not yet been received. In Sudan, garawi is cultivated only to a limited extent, mainly at the experiment station and at military hay farms, two cuttings of hay being secured there each season under irrigation. The seed was brought to Sudan from Egypt, where it is also cultivated to some extent under the same name. It is probable that it is the grass that all writers on Egyptian botany have called Andropogon halepensis. [Cir. 125] 3 4 SUDAN GRASS. The exact nativity of garnwi is slill a matter of doubt, nor is it clear that genuine Androjwgon halepensis occurs in the same region. A few phints of garawi lived over the winter of 1911-12 at Gaines- ville, Fla., without, hoAvever. forming any rootstocks. The second variety was received on December 2, 1009, from Dr. L. Trabut, Algiers, xVlgeria, and given S. P. I. No. 2i>?m. Dr. Trabut's original notes are as follows : T!iis griiss is vigorous but not stoloniferous and would 1)0 intoresting for bybridiziition with soi-ghuui. It is moderately good forage like .Jobusou grass, but has the advantage of not stooling (i. e., suckering). This variety is per- ennial bere and pi'oduces many seeds. Under the conditions in the United States this variety has behaved purely as an annual. In further correspondence with Dr. Trabut, he writes that he believes this grass to be common in Africa and that he has received it from the arid regions between Algiers and Senegal. The two varieties are quite distinct from each other and the name " Sudan grass " has been given to S. P. I. No. 25017 and " Tunis grass " to S. P. I. No. 26301. Botanically, they are both to be con- sidered varieties of Aiidropogon sorghum and not of Andropogon halepensis. as the three knoAvn varieties of the latter all have vigorous underground rootstocks. Trials at numerous places have demon- strated that Sudan grass promises high value for hay, especially in the semiarid regions where no perennial grass has thus far been found suited to the conditions. Indeed, it is not too much to predict that it is there destined to become the leading grass for hay production. Un- der more humid conditions Sudan grass has also succeeded admirably and it will probably replace the foxtail millets to a large extent, as it produces better hay and usually larger yields. Tunis gi\ass has not as yet been widely tested, owing to lack of seed. It is slower in start- ing growth and less tall than Sudan grass. As it shatters its seed very readily it is likely to be of only limited usefulness unless this character can be changed. Sudan grass has been tested most carefully in Texas (fig. 1) and at Arlington Farm, Virginia, but at least one year's trial has been made at many places in the Great Plains and at various agi-icultural experiment stations. There is still much to be learned in regard to the crop, but the data at hand indicate approximately the best methods of culture. Sudan grass is a sorghum and lequires practi- cally the same temperature conditions as that crop. It is, however, earlier than any sorghum yet knoAvn and will probably mature in jNIontana and North Dakota, as it ripened seed in 1912 at Brook- ings, S. Dak. Individual plants of Sudan grass under favorable conditions will attain a height of 8 to 10 feet and may possess 20 or more stalks to a plant. Tlie stems seldom become larger than a lead pencil, even ICir. 125] SUDAX GRASS. in tlie largest plants. Broadcasted or drilled the height averages 3 to 4 feel, and the stems are much finer. The stems are mostly iin- branched, strictly erect, and decidedly leafy, very much more so than Johnson grass. The sugar content is small, but enough to give a decided sweetish taste. The flower cluster is loose and open, pyram- idal in form, and G to 12 inches long. There is practically no shatter- ing of the seed at maturity. SEEDING SUDAN GRASS. Sudan erass mav be sown broadcast, drilled, or in cultivated rows. Where there is sufficient moisture, broadcasting or drilling is prefer- FiG. 1. — Fields of Sudan gras« at Dalhart, Tex. Oii the riglit, uucultivated rows; ou the left, broadcasted. able; otherwise the grass is likely to be coarse. In seeding this way 3 pecks of seed to the acre should be used. Under conditions of light rainfall Sudan grass is probably best sown in cultivated rows, though excellent results have been secured in dry regions from broadcasting. In rows 3G inches wide, 4 pounds of seed to the acre are sufficient, even with rather thick seeding, which is recommended when grown for hay. For seed production much thinner seeding has given excellent results. It is sometimes practicable in humid regions to sow in 18-inch rows and cultivate. This is especially desirable where the land is very weedy. The grass grown under such conditions does not become too coarse, and, furthermore, the dense shade kills out the weeds. Five pounds of seed to the acre should be used when thus sown. [Cir- 125] 6 SUDAN GRASS. FEEDING VALUE OF SUDAN GRASS. All reports agree on the high palatability of Sudan grass, either green or cured. At Chillicothe, Tex., the farm horses even ate readily the straw from which the seed had been thrashed. Until feeding experiments can be conducted no definite statement of the compara- tive feeding value of this grass can be made. Table I shows the analyses of a series of hay samples cut at various dates at Arling-ton Farm, Virginia, in 1912. Perhaps the most interesting feature shown is the close comparison of the mature straw with hay cut at earlier stages. Table I. — Analyses of Sudan grass grown at Arlington Farm, Virginia, in 1012, cut at various dates in different stages of maturity. Cut Aug. 7. Cut Sept. 1, before heading. Cut Oct. 1 ; seed was fully mature. Substance. Before heading. Heads just ap- pearing. Just be- ginnmg to bloom. In full bloom. Moisture Per cent. 4.13 6.61 1.72 7.75 30.68 21.82 27.29 Per cent. 3.54 5.55 1.39 6.06 31.94 24.01 27.51 Per cent. 3.46 5.02 1.23 5.16 33.23 24.70 27.20 Per cent. 3.51 5.64 1.27 4.66 35.12 24. .51 24.79 Per cent. 4.82 7.12 1.49 5.63 34.30 23.38 23.26 Per cent. 4.38 Ash 5. .59 Ether extract 1.48 Protein 4.19 Crude fiber 34.44 Pentosans 26.70 Undetermined 26.70 EXPERIMENTAL TRIALS WITH SUDAN GRASS. Owing to the fact that Sudan grass came from a dry tropical country and that the quantity of available seed was very small, the preliminary tests were all made in Texas. The remarkable adapta- tion of the grass to Texas conditions led to its being tested in 1911 at Arlington Farm, Virginia, and in various Southern States. At tlie former place it succeeded beyond expectation, so that seed of it was sent in the spring of 1912 to many experiment stations with the request that it be tried, but for various reasons comparatively few stations made a test. The reports of these trials are given later in this circular. As most of these tests were very small the results can only be regarded as indications of its j^ossible value. In most cases the grass was seeded in cultivated rows, under which condition it is usually too coar.se for hay of high quality. By thick planting, how- ever, this difficulty is easily overcome. Practically every test of the grass made in the semiarid regions from South Dakota to Texas has given remarkably favorable results. There is scarcely room to doubt the very liigh value of the grass for this portion of the United States. A single test in eastern Oregon also gave very promising results, so the grass is doubtless adapted [Cir. llT.] SUDAN GRASS. 7 to Columbia Basin conditions. Its wide adaptability to the climatic conditions of the United States east of the Rocky Mountains is note- worthy. IVliile the original stock showed little variability, the grass has crossed very readily with sorghums, so that it is possible to select various hybrids differing especially in leafiness and date of maturity. Presumably all of the variants are due to crossing, but no isolated areas of the original seed have yet been grown to determine whether other factors are operative. In the humid regions the results are not so uniformly satisfactory, and the future of the grass east of the 100th meridian can not be forecasted with confidence until much further evidence is available. A few packages of seed were also sent in 1911 and 1912 to farmers for practical trials. The reports of several of these trials are cited as indicative of the value of the grass, and some of them are valuable as suggesting critical experiments. RESULTS OF TESTS AT CHILLICOTHE. Sudan grass was first tested at Chillicothe, Tex., in 1909, a single row being grown and all the seed saved. In 1910 this seed was planted in 30-inch rows on seven-fifteenths of an acre of land. Though the season was exceedingly dry it grew to a height of 4 to 4| feet. A small portion of the plat, one-fifteenth of an acre, was cut for hay and yielded two cuttings. From the remainder, 134 pounds of seed were secured in two pickings,, which is at the rate of 335 pounds per acre. In 1911, plats were planted June 1 on newly broken sod land, from which two cuttings of hay were secured, each larger than the one cutting of German millet grown alongside. The total rainfall from April 1 to November 1 was 14 inches. The drought conditions of the season were such that both milo and kafir produced only about one- fourth of a normal grain yield. I During the season of 1912 more detailed results were secured. Four cuttings of hay Avere obtained from a one-tenth acre plat, drilled on April 26 at the rate of 3 pecks of seed per acre. The date and amount of each cutting are as follows: Pounds. June 22 214 July 17 181 August 20 - 305 October 14 180 Total . 880 [Cir. 125] 8 SUDAN GRASS. This jdeld is at the rate of 4.4 tons per acre. During this period the rainfall was as follows: IiK-hos. April 26 to 30 1 0. OS May .52 June 4.69 July 1. 39 August , 3. 35 September 2. 92 October 1 to 14 1.97 Total 15.52 Two acres were also planted on April 20 in rows 36 inches apart. This crop grew to an average height of 6 feet 4 inches and was har- vested for seed on August 3, 96 days after planting. It was a little overripe and probably 10 per cent of the seed was lost by shattering. The actual seed saved from the 2 acres was 708 pounds. By Septem- ber 20 the grass was again about 18 inches high and beginning to head, when it was cut for hay in order to plow the ground. The yield was estimated at about 1,000 pounds per acre, but it was not weighed, owing to rainy weather. Two fields of Sudan grass were grown for seed under contract with two farmers in the immediate neighborhood of ChiUicothe. One farmer planted 12 pounds of seed on 13 acres in 42-inch rows and secured a yield of about 10 bushels per acre. The second farmer planted 4 pounds of seed on 2 acres in 42-inch rows and harvested 1,285 pounds of clean seed, or 15.3 bushels per acre. The seed grown on the experiment farm weighed 40 pounds to the bushel ; that grown by the first-mentioned farmer, 44 pounds, and by the second, 42 pounds per bushel. In contrast the seed grown on the experiment farm in 1911 weighed but 32 pounds per bushel. RESULTS OF TESTS AT ARLINGTON FARM. At Arlington Farm, Virginia, Sudan grass was tested in 1912, both broadcasted and in 18-inch rows (figs. 2, 3, and 4). The broad- casted plats were sown on June 3 at the rate of 10 pounds of seed to the acre. The broadcasted stand was not perfect, some comparatively sterile spots being almost bare of Sudan grass and occupied by pigeon grass. The crop in these plats grew to an average height of 5 feet. One twentieth-acre plat cut for hay on August 23 yielded 280 pounds, or at the rate of 2.8 tons per acre. The second growth on this plat was 30 inches high and was beginning to head on September 20. This grew to a height of about 3 feet, but the seed was not nuiture when killed by frost on November 4. Nine plats of one-twentieth of an acre each were cut for seed on September 20 and yielded, on the [Cir. 125] SUDAN GRASS. average, pounds, or 3.3 bushels per acre, only one-fourth of the quantity secured from the plats in rows. Fifi. 2. — A broadcasted field of Sudan grass at Arlington Farm, Virginia, 1912. Eight plats of one-twentieth of an acre each were planted in 18- inch rows on June 3 at the rate of 5 pounds seed per acre and culti- ■ ^lgl^^» ■■- '•■- ^^MJ^^^i^^^'f^thi iiiiiiife^i^ I^I^J^Ij^^^^^^^aE^^^^^^^M KJ^9k|^^^^WEA4 f yiiy^JIB^j^jyR JfciJfesiMBB^ CbBk^^^^^S ^^MA^^^^'V^^^^^^^^^^^IP^^^^I^^^ BQlwH|[|kw?^^^^iKjPE 9SBS^sm&S^^^^Ss^^^fl^^SK^^Mi^^'^SSSSii^^^^^^^^f3i^^^ ^ iS^^B!^v'^^aK!lQHI^H^^H MTE^^^S'VJiBJ^B^KlB^^B^^B^K¥j^^>^J^^^^^BfilBBBffK!k^^^^^^^B8f » >/^(>ir3^^-n^^«rH -^r vV>:^^--.J^M/%^- xl"ilV^^^ ^■^/;,.4X .^- \lvw §snniaKP^^^H^im%iSB^i'«KKSg%s , \:M S^~ O^M^ iv^\-^i\:''^^Q-'a:^-Vi^f- i.j"^4i^^^ kr-^T-iil 4jfmmM?^^^^jtw ki^^ ^;,V._.^> :^E.^ "^mmmmsBmmmmR m^smmm^s^i f!S ""■■ ■' '^'S^^^^^^S Fig. 3. — Rows of Sudan grass at Arlington Farm, Virginia, 1912. Each row is grown from the seed of a single plant. The three rows on the right are typical Sudan grass. vated twice. Two of these plats cut on August 23 when fully headed and about 7 feet high yielded, respectively, 284 and 347 pounds per 2)lat, or at the rate of 2.8 and 3.5 tons to the acre. This could have 82199°— Cir. 125—13 2 10 SUDAN GRASS. been cut as early as August 10 with a veiy slightly smaller yield. The second growth on these two plats was over 3 feet high Avhen killed by frost on November 4. The remaining 8 plats were harvested for seed on September 20 and yielded an average of 23 pounds each, or 4G0 pounds per acre. Practically no seed was lost by shattering. The second growth in these plats w^as about 1 foot high when killed by frost on November 4. A late seeding was made on August 7 in rows, and this was 48 inches high and fully headed when killed by frost on November 4. There can be no doubt that by seeding not later than June 1 two full cuttings of Sudan grass for hay can be obtained each season in Fig. 4. — Sudan grass at Arlington Farm, Virginia, 1912. This is another view of the right-hand row shown in figure 3. The tall plants in the background are hybrids be- tween Sudan grass and some variety of sorghum. Virginia. The grass has shown much stronger growth in cultivated rows than when broadcasted, but it still remains to be determined which method is most desirable. The seed grown at Arlington Farm in 1912 weighed 36 pounds per bushel. Mixtures of Sudan grass with cowpeas and with soy beans were also tested (fig. o). A one-tenth acre j^lat was broadcasted on June 11 with 3 pounds of Early Black cowpeas and 2 pounds of Sudan grass. This was cut for hay on September 6 Avhen the Sudan grass was in bloom and the first pods of the cowpeas were fully grown. The grass was 6 to 8 feet high and the cowpea vines were of about an equal length. The plat yielded 925 pounds of cured hay, about one- [Cir. 12.';] SUDAN GKASS. 11 fourth being cowpeas. This is at the rate of 4.() tons of the mixture per acre. In an adjacent one-tenth acre plat Johnson grass and cowpeas were seeded at the same rate; that is, 2 pounds of Johnson grass and 3 pounds of Early Black cowpeas in place of Sudan grass (fig. 5). The yield of the mixture was 561 pounds of air-dry hay, or 2.8 tons per acre. A similar mixture of Sudan grass and Arlington soy beans, a twin- ing variety, was sown the same date, using 3 pounds of soy beans and 2 pounds of Sudan grass. About one-fourth of the mixture was soy beans, which twined about the grass to a height of 4 to 6 feet. When Fig. 5. — riats at Arlington Farm, Virginia, sliowing mixtures of Sudan grass and cowpeas (rigbt) and Johnson grass and cowpeas (left). cut on September 6 the Sudan grass was in bloom and the soy-bean pods w^ere about half grown. This mixture cured more readily than the cowpea mixture and was superior in physical quality. The yield was 888 pounds of cured hay, or at the rate of 4.-1: tons per acre. Figure 6 shows a stand of Tunis grass planted in rows at Arlington Farm for comparison with the Sudan grass shown in figure 3. TESTS AT MISCELLANEOUS EXPERIMENT STATIONS. TEXAS. At the San Antonio (Tex.) Field Station Mr. S. H. Hastings tested Sudan grass in 1911 and 1912 and reported as follows: From the growth of the plat tested in 1911 this apijears to be the most prom- ising grass that has been grown at the experiment farm. Two plats were [Cir. 12.5] 12 SUDAN GRASS. planted, one where it could be irrijiated and the other without irrigation. The plat not irrigateer acre. The second cutting was made on October 10 and gave a yield of 3.11 tons i»er acre, making a total for the season of G.OO tons per acre. At least three cuttings would have been secured had it been seeded earlier and the first cutting not allowed to seed, which would have increased the yield materially. In 1912 we put in a planting of Sudan grass March l."», without irrigation, and the yield from two cuttings was 5.0(5 tons per acre. Sorghum planted under the same conditions gave a yield of 4.68 tons. Fig. 6. — Rows of Tunis grass at Arlington Farm, Virginia, l'J12. Note the very uiucli thinner appearance of this grass as compared with the Sudan grass in figure :!. At College Station, Tex., a- test was made in 1912 by Mr. A. B. Conner, who sent in the following report : Planted May 15 on one-fifth acre plat in rows 3 feet apart. Germination was fairly good, but stand not as uniform as desirable. Grass made very vigorous growth up to July 1. On July 7 was just coming into full boot. On July 15 it was in full head at a height of 7 feet and presented a very vigorous apiiear- ance. On August S the plat averaged 7 feet in height, and on account of tlie irregular stand each plant had put out a number of culms. Some were noted with as many as 40 to 50. Plants were very leafy to the top, showing superi- ority in this respect to Johnson grass. Harvested August 14 for seed and gave a yield of 57 pounds of thrashed seed, A second growth, which was i)r(Mluced without any rainfall, the season being exceptionally dry and not enough rain to produce a second growth on sorghums, attained a height of above 5 feet and [Cir. 125] SUDAN GRASS. 13 was harvested for seed October 23, yielding ouly S i)ouuds. The total yield of seed from the one-fifth acre plat was 05 pounds, or at the rate of 325 pounds per acre. Seed tested 19 pounds per bushel, which gave the equivalent of 17+ bushels per acre. A test at Dalhart, Tex., gave very promising results, thus reported by Mr. ^Y. D. Griggs: Two one-tenth acre plats of Sudan grass, one broadcast and one in rows, were seeded May 2, 1912 (tig. 1). The plat in rows failed to give a good stand and it was reseeded May 21. It was intended that these plats be harvested for hay, but owing to the local demand for seed among farmers it was decided to let the grass mature and harvest it for seed. Both plats were harvested September 7 and gave a total yield as follows: Broadcast. 545 pounds; in rows. 552 pounds of hay per plat. The former yielded 40 i)ounds of seed; the latter, 54 pounds. It is estimated that 25 i>er cent of the seed was lost In harvesting. CALIFORNIA. The following is a report on a trial at the Plant Introduction Field Station, Chico, Cal., by Mr. Roland McKee : Two rows of Sudan grass, one 75 feet long and the other 150 feet long, were grown at Chico in 1912. It was grown on good loam soil and given Irrigation. A fine growth was made, and without question this is the most jiromising grass for growing imder irrigation in the Sacramento Valley that has yet been tried. The number of cuttings of hay that it is possible to secure was not determined, as with both the plantings a seed crop was allowed to mature, but it seems probable that three good cuttings of hay can be made. One of the plantings was allowed to produce a seed crop from the first growth. This was harvested late, but still a good hay crop was produced after that date. The other plant- ing was cut for hay shortly after it came into bloom. A good crop of hay was secured and after that date a seed crop was matured. The following data give some idea of the growth of this crop: May 2 Itow 75 feet long; sown. July 9 In full bloom and 48 to 72 inches high. July 15 — Cut for hay. August 29 Second growth GO to 90 inches high, KQveniber 1 A good crop of seed was ripe. May 13 Row 150 feet long; sown. July 9 In first bloom and 30 to 40 inches high. August 20 First seed rljie. August 28 70 to 80 inches high. September 14_^ Cut for seed; 18 pounds secured. Yield about 40 liushels per acre figured on b;!sis of 4-foot rows and 30 i)ounds of seed per bushel. November 3 Second growth 3 to 4 feet high. SOUTH DAKOTA. A small test made in 1912 at Brookings, S. Dak., is thus reported by Mr. Samuel Garver: Three rows, 36 inches apart, each 8 rods long, were planted Aiiril 30. The grass grew very slowly during the cool spring, being only 3 or 4 inches high on June 10, and was not injured meantime by three or four frosts. At this time warm weather began and the grass grew rapidly, maturing its seed [Cir. 125] 14 SUDAN GRASS. September 16. The nctual amount of seed harvested from the three rows was 1S.5 pounds, which is at the rate of 678 pounds per acre. June 12, two rows. 36 inches apart, each S rods lon^. were planted. One vow was cut for hay September 16, when 6^ feet higli, but it should have been cut earlier. The yield of this row was 110.5 pounds, or at tlie rate of 6.08 ions per acre. The second row was left for see. The jilanting made on June 3 consisted of al>out onc-fdurth of an acre. It came up well, and there was a good stand to begin with, but owing to various att-idents during the season there was not more tlian about one-third of a stand left to produce [Cir. l:i5] SUDAK GRASS. 17 seed. The yield of seed was about 3 bushels aud was harvested the first week in Noveuiber. That planted on July 13 did not mature seed before frost, but was ready for cutting for hay after the middle of October. The grass seems very promising for this section, aud it appears from the late seeding that over a large part of Maryland it would be possible to sow the grass after Avheat and harvest a crop of hay that year. The hay was relisheer acre. A few days after being cut it began to grow from the stubble. Having no rain it grew slowly until Sepember 21, 1912. On that date we had our first rain since June, 1912. After that date it began to grow quickly until November 6, 1912. November 1, 2, and 3 we had a severe frost which did not seem to hurt it at all. November 6, a dry blizzard came down on us. Being afraid I was go- hig to lose the grass, I cut it and tied it, still green, in bundles and hauled it to the barn, where it cured and made me plenty of tine feed, it being 4 feet high at the time of cutting. If I had cut this at the proper time to make good feed, i. e., when the seed was in the boot, I could have cut it four times instead of twice, but I was anxious to save the seed. The terrible drought blasted the first crop and the frost prevented the second from maturing.' I exhibited at the Boerne fair, September 6 and 7, 1912, a bunch of this grass that measured 10* feet high. Agents of the Agricultural Department, College Station, Tex., who acted as judges at the Boerne fair, were astonished. They told me they had the grass drilled just as I had drilled mine and it only grew 4 feet high for them. Dr. W. O. Langdon, of Hutchins, Tex., grew Sudan grass in 1911 and again in 1912. He thus reported his 1912 results, under date of August 18: The seed received this spring was planted In a little piece of ground in rows about 30 inches apart. It made a fine growth of an average of G feet. Much of it was nearly 8 feet high. When .seed was ripe, about xlugust 1, I cut it. It im- mediately Legan a second growth and is now nearly 3 feet high. The ground is covered with young plants from shattered seed. I think it is the greatest for- age plant ever introduced into this section and that it w ill be worth millions to 1 The blasting of the seed referred to by Mr. McCarthy is perhaps due to the work of the sorghum midge, which attaclis Sudan grass, like other sorghums. ICir. 125] SUDAN GRASS. 19 Texas and other seruiarid sections. All stock eat it ravenously, as it is very sweet. Tlie inclosed pictures will give you some idea of it, although it must be seen to be fully appreciated. In a supplementary rejDort dated January 20, 1913, he added : In answer to your letter will infoi'm you that the second growth of Sudan grass attainetl a height of at least an average of G feet. I cut and stored the hay from the first cutting, hoping to get the seed thrashetl out, but did not suc- ceed in doing so. I did not cut the second growth, but am sure it was as heavy as the first crop. I am sorry I can not give you more definite information as re- gards the amount of hay and seed produced, but have been very much " under the weather " for some months and so things have gone slack. Of one thing I am certain — there is no better forage plant for this section. It is a wonderful pro ducer, very drought resistant, makes a most i-emarkably sweet-smelling hay with a very difl:"erent odor from Johnson-grass hay, and will never become a pest. Mr. J. R. Stegall, of Detroit, Tex., Avrote concerning his 1912 expe- rience as follows : I sowed the Sudan seed that you sent me, half in bottom on strong land and the other half on sandy, thin land. The results on both are most wonderful. I prepared a good seed bed by breaking deep and then harrowed both ways. I sowed seed April 25, making the first cutting May 20. I have cut the twentieth of every summer month, June, July, and August. I am cutting to-day (August 20, 1912). The yield is larger every cutting, as it stools out' from the root. The piece in the bottom I sowed by the side of a small spot of Johnson grass. I have cut the Johnson grass twice and the Sudan grass four times. Stock eat Sudan grass hay with more relish than Johnson grass hay, as the texture is not so coarse. This Sudan grass is the most wonderful thing in the way of hay I ever saw. I have had a great many applications for seed, but I have none to sell. In a postscript Mr. Stegall added : I neglected to mention we have had quite a long, dry time; no rainfall. Not all the crops were damaged. The Sudan grass has resisted the drought, the grass on the bottom land standing better than that on the sandy soil. In a subsequent report additional data were given as follows : You sent me 2 poinids of seed. I sowed it on one-eighth of an acre. I got 400 pounds of nice, cured hay at each cutting. Some of my neighbors told me I sowed it too thick, but for cutting in milk I am satisfied I sowed it about right. When I turnetl my cattle, hogs, and horses in the field, I had goobers, peas, corn, and crab-grass, but they would not eat any of these until the Sudan grass was completely consuuied. Stock love it better than anything else they can get to eat. I had a small patch of Johnson grass right by the side of the seed I sowed in the bottom. I cut the Johnson grass twice and the Sudan grass five times. The weather was dry and I got a nice cure every time I cut. Stock would eat the Sudan grass before they would the Johnson grass. Mr. W. W. Price, Mount Pleasant, Tex., recorded his experience as follows : On April 3 I sowed 5 pounds of Sudan-grass seed on one-half of an acre of light, sandy land. The grass reached a height of 7 feet and matured 64 days from the date sown, the cutting yielding 1| tons of hay. [Clr. 125] 20 SUDAN CRASS. The experience of Mr. H. N. Montgomery, of Austin, Tex., is thus reported : We I'laiited ilie Sudan i;r:iss broadcast, about like oats, on a rich 1)U) rather droughty piece of land (black waxy) which, however, had been well prepared and was in good condition. Since planting. May 7. we have had one good rain, June 17. The grass had withstood the drought well and made a very rapid growth, attaining an average height of 5 feet. Although very similar in ap- pearance to Johnson gi'ass, I consider it far superior as a forage crop, as it is much more bunchy, putting out more stems anlants have the same ap- pearance as those affected by drought, except that green and withered stalks may be seen arising from the same stool. In a bad at- tack these withered stalks become light in weight, fall over, and con- tain practically no juice. If a stalk is s])Iit open at the base durmg the early stages of the disease, the fibro vascular bundles may be seen as reddened streaks, which gradually increase in diam- eter, spreading outward into the neighboring tissue. Later, white blotches consistmg of dead cells filled with air may be found in this reddened area. (See fig. 1.) The disease advances upward toward the tip, the tissue finally becoming dark to blackish m color. If the stool is examined it will be found that the discoloration of the stalk extends downward into the cutting and may be traced from here into other stalks which outwardly seem perfectly normal. The fungous spores are usually I Edgerton, C. W. Tho red-rot. of sugar cane, Louisiana Agricultural Experiment Station, Bulletin 133, 22p.,4pl.,1911. ■ Butlor, E. J. Fungous diseases of sugar cane in Bengal. Memoirs, Department of Agriculture, India, Botanical Series, v. 1, no. 3, ryi p., U pi., 1906. [Cir. 126] Fig. 1.— Red-rot of sugar cane, showing white spots (x) in center of red blotches. (After Lewton-Brain.) FUNGOUS DISEASES IN SHIPMENTS OF SUGAR CANE. 5 formed i:)n the dry sugar cane at the nodes or in sunken areas on the internodes. There has been considerable doubt among investigators as to the manner in which the disease is distributed. Some attribute its spread entirely to the agency of cane borere. Dr. Edgerton states that in Louisiana the infection takes place chiefly through the agency of the moth borer (Diatraea saccharaUs) . In an examination of 55 borer cases, he found that 25 were infected with red-rot, 25 not diseased, and 5 doubtful., Mr. Butler concludes from his observations that the disease is carried in the sets themselves. He found that the yellow Bourbon cane which came from infected districts gave diseased shoots, while the same variety obtained from disease-free regions and grown adjacent to the former plants produced healthy shoots. It would seem from this experiment that the use of diseased stalks for cuttings is the chief source of infection, although cane borers undoubtedly spread the disease to some extent in those fields already infested with spores from the previous season's growth. While the disease may be detected in many cases by inspection, an absolute guaranty of disease-free cane could not be given. ^Ir. Butler also says that — Unfortunately, even with the greatest care it is not possible to eliminate diseased sets entirely, for I have obtained cultures of the fungus from sets passed by a particu- larly careful man, and the reddening does not always show at the cut ends. And in speaking of the red discoloration of the tissue, which is considered the chief diagnostic character, he states that — There is no peculiarity in this discoloration that I have been able to find which serves to distinguish it in all cases from the reddening due to other causes. Control. — Selection of resistant varieties -and selection of "seed" are recommended for controlling the red-rot. ILIAU. Iliau {Gnomonia iliau Lyon) is a disease apparenth^ peculiar to Hawaii, where it is said to have been kno\v^l since the beginning of the sugar industry. Although the disease has been recognized for many years, it was not until 1912 that the entire life history of the causal organism was worked out and published.^ Dr. Lyon considers that far greater losses to the sugar planters in Hawaii have occurred from this disease than that caused by all other fungoid diseases combined. The following instances of loss are quoted from his bulletin: The manager of one plantation estimated that fully one-half of the shoots in a 200- acre field of 4-monlhs-old plant cane had been killed out by iliau. His estimate was 1 Lyon, H. L. Diau, an endemic cane disease. Report of Work of the Experiment Station, Hawaiian Sugar Planters' Association. Pathological and Physiological Serias, Bulletin 11, 32 p., 10 fig., 1 pi., 1912. [Civ. 120] CIKCULAE NO. 126, BUEEAU OF PLANT INDUSTRY. made at a time when the disease was a( its height in the field. At a later j)eriod, during warm weather, this cane recovered rapidly, the stand being greatly augmented by new shoots. When the field was finally harvested the yield in sugar was about 20 per cent less than the average amount which this field could be expected to produce. The loss in this case amounted to a little over a ton of sugar on each acre. As no other disturbing factors appeared during the growth of this crop, the loss can very reasonably be ascribed en- tirely to iliau. We have seen several cases where iliau has killed all of the stools over small areas of plant cane and one case where it effectually destroyed the cane over an area of 15 acres. We have seen many in- stances where the disease has killed from one- fourth to one-half of the primary shoots in large fields, our estimates for several cases being based on careful counts made in various parts of the af- fected areas. Iliau is a Hawaiian expression meaning "tight skin" or "liideboimd," having reference to the manner in which the leaf sheaths are funnly held together around the stem by the fungous mycelium. Any attempt to remove the leaves is met with considerable resistance. The f ung-us gains an entrance into the tissue of the cane at the leaf bases usually below the surface of the ground and from here makes its way into the leaf sheaths, working up- ward and then inward. The leaves whose sheaths are diseased soon die but do not fall off, as they are tightly cemented to the sheaths witliin. (See fig. 2.) On young cane this tight case prevents the shoot from elongating and thus eventu- ally kills the terminal growing point. If the stem or shoot outgrows the mycelium, i. e., before the fungus has penetrated the inner sheaths, the disease does little harm. The fungus can make material progress only in young and tender tissues, and even if it has entered the stem its gi'owth is checked by the normal hardening of the tissues. The d{>ad l(>af sheaths are pinkish brown in color and the rind, if any growth has been made, is deep bluish gray. The imperfect form of this fungus (Melanconium) is the one most commonly met with on the diseasiul shoots. It ])elongs to the same ICii-. 12G1 F:g. 2.— Iluui disease of sugarcane after an attempt to remove the leaves. (After Cobb.) FUNGOUS DISEASES IN SHIPMENTS OF SUGAp CANE. 7 genus as the fungus causing rind disease of cane. There should be no confusion in distinguishing the two diseases in the field, the one being a vigorous parasite attacking only young shoots, while the other is a saprophyte found on cane killed from some other cause. The spores are formed in pustules usually within the tissue of the stem or the inner leaf sheaths and are not released until the stem disintegrates. So long as the shoot remains dry the pustules remain intact, but when moistened the sheaths loosen and the spores are freed. The disease flourishes only (hn-ing damp, cool weather and then only on young shoots. Fields in which the cane is more than half grown are not affected, the disease being confined in that case to the young shoots or suckers. Inspection of seed sugar cane for this disease would be inadequate, since the disease is usually not evident on mature cane. S]:)ores of the fungus might be carried on the surface of the lower portions of the stalks which had outgrown the disease or might find lodgment on healthy cane coming from infected areas and the disease be thus carried. Control. — Thorough preparation of the soil previous to planting is recommended in controlling the disease, since an exposure of the spores and mycelium to the sun for a fe\y hours effectually destroys them. GUMMING. Gumming {Bacterium vascular^im (Cobb) Greig-Smith) , a disease of the fibrovascular bundles of sugar cane, was first described in 1893 by Dr. N. A. Cobb ^ from the Clarence River district of Australia. It is also present in Mauritius, Java, and Brazil. TVTiile this trouble has so far been easy to control in infected regions it might become far more serious if introduced into a new habitat. The disease is characterized ^ by a dying of the cane tops, a rotting of the tissues at the base of the "arrow," and the production of a yellowish gmmny substance in the bundles of the stem. Gmmning is propagated through diseased sets. The bacterial organism ad- vances in the bundles of the stem along with the growing plant. In severe cases the plants are dwarfed, attaining in a year's growth only a third or half the size of healthy plants. Occasionally the young plants are killed before they have had time to push through the ground. If a diseased stalk is cut crosswise, a sticky yellowish sub- 1 Cobb, N. A. Diseases of the sugar-cane. Agricultural Gazette of New South Wales, v. 4, pt. 10, p. 777-833, 46 fig., 1893. 2 Cobb, N. A. Third report on gumming of the sugar eano. Report of Work of the Experiment Station, Hawaiian Sugar Planters' Association. Division of I'utliology and Physiology, Bulletin 3, 40 p., 12 fig., 2 pi., 1905. [Cir. 126] 8 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY, stance exudes from the bundles in the form of tiny drops. (See fig. 3.) This is the most characteristic symptom of the disease. Ordinary inspection, however, is not sufficient to detect this disease in cases where the cane is only slightly attacked. Control. — Selection of cuttings and growing of resistant varieties are recommended. SMUT. Smut ( Ustilago sacchari Rabenhorst) is rather widely distributed in cane-growing countries, but usually does little damage. Mr. Butler describes the disease as follows : Affected plants are known by the production from the growing apex of a long whip • like dusty black shoot, often several feet in length and much curved on itself. (See fig. 4.) This shoot is an entirely abnormal growth, devoid of leaves, slender, and flexible. It must probably be taken to represent an abnormal floral shoot. In its earlier stages it is covered by a silvery white thin sheath which soon, however, ruptures, exposing a dense black dust consist- ing of the spores of the fungus. From the upper portion of the affected cane no sec- ondary shoots arise, but from its lower part they are fairly abundant, being all in their turn attacked and prolonged into spore-containing organs. In this way it becomes evident that the whole plant is affected. The tissues of the cane below the abnormal shoot con- tain the filaments of the smut fungus. These are found between the cells, not in them, as is the case with Col- letotrichum falcatum. Diseased canes are poor in sugar and worthless.' The disease may be transmitted through spores coming in contact with healthy plants or through diseased sets. The latter means is probably the most frequent. Control. — Care should be exercised to pre- vent the spread of spores through tools, cloth- ing, packing, manure, etc., and only cane from healthy plants should be used for the sets. SEREH. Fig. 3.— Gumming disease of su- The causc of serch is unkuown, although it gar cane, showing gummy exu- ^ ^^^^^^ recoguizcd as far back as 1882 in dations on cut surface of cane. ^ (After Cobb.) Java, whcrc it has done much damage.- Dr. Lyon reports it to be of rare occurrence in the Hawaiian Islands, having been first recorded in 1908. 1 Butler, E. J. Op. cit. 2 Di'crr, Noel. Cane Sugar. .Xllrinchain (M;uichesti'r), p. 14") 147, fii;- T2. 191 1. ICir. 12(;j FUNGOUS DISEASES IN SHIPMENTS OF SUGAR CANE. 9 Mr. Noel Deerr states that — In the typical form of "sereh" the stool of cane consists of a number of short stalks with very short joints; the buds, especially those below, sprout, whereby results a bundle of short stems hidden in a mass of leaves. The whole stool bears a resemblance to lemon grass (Andropogon schoenanthus) , the Ja\anese term for which is "sereh." In a second type one or two stalks may grow to a fair size with very short joints in the upper part; above all is a fan-shaped leaf crown; many of the eyes, especially those below, sprout and form small branches. * * * x red coloration of the fibrovas- cular bundles is a characteristic of "sereh." This coloration is most pronounced in the node, but often appears in the internode in the form of a red stripe. This appear- ance is quite distinct from the red patch with white center characteristic of the red-rot of the stem. A point of great interest with this disease is the difference of opinion as to its infective nature. The disease certainly spread from district to district in Java, but, conversely, healthy sticks planted in an affected field remained healthy. Whether infectious or Fig. 4. — Smut disease of sugar cane. (After Butler.) not, the disease was found to be hereditary; that is to say, canes planted from sound, healthy seed gave healthy canes, but tops derived from sereh-struck canes became equally infected. THE PINEAPPLE DISEASE. The pineapple disease {Thielaviopsis jMradoxa (De Seyn.) V. Holm) was first studied by Dr. Went in Java m 1893. Since that time it has been found also in Mauritius, British Guiana, the West Indies, Hawaii, and was reported from Louisiana in 1910. Mr. Butler reported that up to 1906 he had fornid the pineapple disease in Bengal on only three occasions and these were on cane which had been imported from Java and Mauritius for seed. This disease is one which chiefly attacks the planted cuttings, the fungus being a woimd parasite. Considerable damage is caused wherever this disease is 90341°— Cir. 126—13 2 10 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY. prevalent through a failure of the "eyes" to germmate. Dr. Cobb considers the pineapple fungus to cause the decay of more cane cuttings in Hawaii than that produced by any other one cause and has estimated that in diseased fields the loss varies from 1 to 25 per cent of the plantings. Fortunately the disease has not become widely distributed in the United States, having only been reported from a few localities in Louisiana.^ The fungus effects an entrance into the cane at the cut ends of the "seed" or at any point along the stalks through wounds of the rind made either by mechanical or insect injury. The disease (fig. 5), marked by pmk or red streaks, progresses rapidly through the central tissues of the cane, eventually kilhng the buds and thus preventing germination. Later, this central diseased tissue turns dark and finally sooty black through the production of macroconidia. This blackened central tissue is what is known as the "pipe" ^ and is one of the chief characteristics of the disease. It is in the younger stages of the disease that the pineapple odor may be detected, being more noticeable in some varieties of cane than in others. These two characters, viz, the "pipe" and the pme- apple odor, are the chief marks by which the disease may be detected. The same fungus causes a disease of pineapples. Control. — Destruction of diseased cane and the immersion of the cuttmgs in Bordeaux mixture shortly after the cane is cut are rec- ommended. In the West Indies Howard also recommends that the ends of the cane be dipped in tar as a further preservative. SCLEROTItTM DISEASE. The Sclerotium. disease (Sclerotium rolfsii Sacc.) is one which at- tacks the leaf sheaths principally, being usually confined to the lower sheaths. It was reported from Louisiana m 1908 and has smce been found in Georgia. The amount of damage caused is somewhat questionable, some clauning that it is confined to the sheaths and does very little harm, while others assert that the fungus penetrates mto the stalk for some little distance and also mto the bud, preventing its germination when used m plantings. This disease is probably the same as that described by Kriiger ^ under the name of red-rot of the leaf sheath and stem and attributed by hmi to Sclerotium sp. He states that the disease is characterized by a brick-red discoloration 1 Edgerton, C. W. Some sufjar-cane diseases. Louisiana Apriiultural Experiment Station, Bulletin 120, 28p.,12fig., iniO. 2 Cobb, N. A. Fungus maladies of the sugar-cane. Report of Work of t he Experiment Stat ion, Hawaiian Sugar Planters' Association. Division of Tathology and Physiology. Bullet in 5, 254 p., 101 fig., 7 pi., 1906. 3 Kruger, Wilhelm. Das Zuckerrohr und Seine Kultur. Magdeburg and Wien, p. 459-464, pi. 14, fig. 2-3, 1899. [Cir. 120] FUNGOUS DISEASES IN SHIPMENTS OF SUGAR CANE. 11 of the sheath, which does not have a definite border. Later, the discolored tissue near the center takes on a blackish, sooty tinge. The Sclerotium disease enters the stem through the bundles at the point of attachment of the leaf and then advances both above and below the node, frequently forming cankers on the stem. (See fig. 6.) Infection rarely, if ever, takes })lace at the internodes. If Fig. 5.— Pineapple disease of sugar cane. The cutting at the right was completely ruined in a few days, although its condition when planted was like that of the sample shown at the left. (After Cobb.) the attacked leaf sheath is pulled off there may be seen a network of fine white shiny threads of the fungus at the base of the sheath. Later, the sclerotia, or resting stage of the fungus, consisting of small round bodies, at first white, then light yellow, and finally brown, arc formed. These sclerotia remain in the tissue of the cane or drop oft" on the ground and are capable of carrying the fungus through con- ditions unfavorable for its growth, such as dryness. Under favorable [Cir. 126] 12 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY. conditions, usually in the spring, these bodies produce the mycelium, which may penetrate those leaf sheaths with which it comes in contact. The disease flourishes during damp weather. Control— RemoY id and burning of the lower leaf sheaths and leaves are recommended. Infected cuttmgs should not be used for plantings. Kriiger states that if infected cut- tings come up at all they seldom ])r()ducc well-developed plants. ROOT DISEASE. Root disease is probably present in all the sugar-cane regions of the world. It is commonly caused by one of the following fungi: Maras- mius plicatus Wakker, M. sacchari Wakker, or Ithyphallus coralloides Cobb. Great losses have been sus- tained from this trouble. Dr. Cobb ^ estimated that in Hawaii failures frequently amounted to 25 per cent and in a few cases as high as 50 or 60 per cent, while a conservative estimate was placed at 10 per cent of the ratoon crop and a somewhat smaller loss on the plant crops. In Louisiana Prof. H. R. Fulton - states: It is a usual thing to find the purple plant cane affected to the extent of 5 to 8 per cent of the stalks, and purple first-year stub- ble, 12 to 18 per cent. In the case of D 74, 1 to 3 per cent and 4 to 8 per cent are usual figures for the corresponding crops. Some of the worst fields seen have been in purple cane and have had 90 or 95 per cent of the stalks infected. These figures are not to h(> taken as representing in any way the loss due to the root disease. Affected canes, although usually small and light, are by no means a total loss; and the fig- ures do not take into account the gaps in stands and the reduction in number of canes per stool that make tip the large loss from root disease. Fig. 6.— Sclerolium disease of sugar cane, showing darkened, sunken areas below the nodes. 1 Cobb, N. A. Op. cit. 2 Fulton, II. R. Root disease of sugar cane in Louisiana. Louisiana .Vgricultural E.xpenment Station, Bulletin 100, 21 p., 8 figs., 1908. [Cir. 12G] FUNGOUS DISEASES IN SHIPMENTS OF SUGAR CANE. 13 The organisms causing root -rot are wound parasites, usually living a saprophytic life in the soil. Wlien the cane is weakened from any cause, such as drought, poor cultivation, etc., the fungus can make an entrance into the roots. The root fii'st shows a reddening of the surface, followed by a brown discoloration, and finally a slow disin- tegration. Plants so affected are weak and dwarfed and easily blown over by the wind or uprooted. The lower leaf sheaths are frequently enveloped hi a wliite weft of fungous threads which cement them tight to the stalk. The disease is nmch more prevalent on ratoon crops than on planted cane, a;S it lives and grows on the stubble year after year. On planted cane the fungus envelops the stalk and soon smothers the buds, thus preventing germi- nation and resulting in a poor stand. Control. — Dr. Fulton suggests the following preventive measures : ( 1 ) Careful cultivation, (2) selection and disinfection of seed cane, (3) plant- ing of resistant varieties, (4) destruction of in- fected trash, and (5) ro- tation. So far as known, no other crop is affected by root disease except .sweet potatoes. ^ This has been reported from Barbados, St. Lucia, and Antigua, of the West Indies. RIND DISEASE. Rind disease {Trichosphaeria sacchari Massee) was once held to be the cause of a great amount of damage in the West Indies and else- where. More recently the rind fungus has been considered by various investigators to be a secondary factor appearing only after the cane has been weakened from some cause or following some parasite such as the red-rot organism. The fungus causes little damage other than hastening the complete destruction of already diseased cane. (See fig. 7.) I South, F. W. Disease of sweet potatoes. West Indian Bulletin, v. 11, No. 2, p. 82, 1911. [Cir. 12G] Fig. ".—Rind disease of sugar cane, showing long, black threads of spores at "c." (After Cobb.) [Cir. 126— B) THE WORK OF THE YUMA EXPERIMENT FARM IN 1912/ By W. A. Peterson, Farm Superintendent, Office of Western Irrigation Agriculture. INTRODUCTION, The Yuma Reclamation Project is characterized by climatic condi- tions peculiarly favorable to the production of a great variety of crops. The rainfall of the region averages only about 3 inches a year and is not a factor in crop production. The climate is warm during most of the year, the temperature for a period of 35 years rangmg from a minimum of 16° to a maximum of 120° F. These conditions are specially favorable to such crops as cotton, alfalfa and alfalfa seed, figs, dates, sweet potatoes, and other crops which require a warm climate and which are well suited to intensive culture. Since the establishment of the Yuma Expermient Farm (fig. 1) on its present site in 1909, experiments have been conducted with a ^i£>-'j?^ j5-,aLAtV'C - Fig. 1.— View of buildings on the Yuma Experiment Farm. view to improving the cultural methods applicable to the above- named crops, to finding superior varieties, and to breeding superior strams of the varieties under test. Much of this work is done in cooperation with other offices of the Bureau of Plant Industry. The offices concerned in the cooperative work in 1912 are the follow- ing: The Office of Crop Physiology and Breeding Investigations 1 Issued May 10, 1913. The Yuma Experiment Farm is located on the Yuma Reclamation Project, 7 miles north of Yuma, Ariz., and adjacent to Bard, Cal. It consists of 160 acres of land, all of which is hrigable. The land was with- drawn from entry in 1909 by the Department of the Interior to be used as an experiment farm. Opera- tions on the tract were begun in the spring of 1910. A farmhouse and an office building were constructed by the Reclamation Ser\-ice; a tool house and a machine shed have been built by the United States Depart- ment of Agriculture. (See fig. 1.) The farm is under the direct supervision of a superintendent detailed from the Office of Western Irrigation Agriculture, Bureau of Plant Industry, and that office furnishes the funds necessary to maintain the farm. On January 1, 1913, Mr. R. V.. Blair succeeded Mr. Peter.son as farm superintendent. |("ii-. 12t;] 15 16 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY. cooperates in the work with figs and dates; the Office of Foreign Seed and Plant Introduction in the work with bamboo, carobs, etc.; the Office of Acclimatization and Adaptation of Crop Plants and Cotton-Breeding Investigations in cotton acclimatization; the Office of Alkali and Drought Resistant Plant Investigations in breeding Egyptian cotton; the Office of Corn Investigations hi testing corn varieties; the Office of Cooperative Cotton Handlmg and Marketing and Paper-Plant Investigations in the work with ramie; the Office of Fiber Investigations in the experiments with hemp; and the Bio- physical Laboratory in the climatological observations. CONDITIONS ON THE PROJECT. CLIMATIC CONDITIONS. In Table I the clunatological observations made at the experiment fai-m in 1912 are briefly summarized, together with a statement of the average weather conditions during the three years 1910 to 1912, inclusive. Table I.— Summary of the climatological observations at the Yuma Expenment Farm, 1910 to 1912, inclusive. Precipit.\.tion (Inches). Item. Jan. Feb. Mar. Apr. May. Jiine. July. Aug. Sept. Oct. Nov. Dec. Total. Average for 3 years, 1910 to 1912 For the year 191^.... 0.22 0.37 0.48 .89 0.18 .26 0.25 .75 0.17 .51 0.37 .20 0.05 .13 0.19 0.35 .33 0.20 0.003 .01 2.886 3.08 Evaporation (Inches). Average for 3 years, 1910 to 1912 For the year 1912.... 3.42 3.96 4.48 5.30 0.92 7.71 6.58 6.40 10.8 9.0 11.25 10.14 10.29 9.44 10.31 9.00 8.27 7. 63 6.47 5.61 4.46 4.19 3.41 2.80 86. 60 81.24 Dah^y Wind Velocity (Miles per Hoxjr). 1910, highest 1911, highest 1912, highest 1910, lowest 1911, lowest 1912, lowest 1910, average 1911, average 1912, average 7.4 5.4 .8 1.5 3.0 3.2 8.5 7.4 1.1 3.5 4.0 5.4 9.3 1.3 1.4 2.7 3.9 9.2 8.4 1.5 1.7 3.8 3.8 9.0 8.1 6.9 2.3 1.9 1.5 4.1 3.3 3.3 6.3 4.4 1.1 1.4 1.0 3.8 2.9 2.6 7.4 6.0 5.1 1.5 1.3 1.5 3.6 2.5 2.3 7.2 3.2 5.9 2.0 1.2 3.4 2.0 2.3 4.6 4.1 4.7 1.2 1.1 .3 2.4 1.7 2.1 9.2 6.7 7.9 1.2 3.7 2.5 3.0 6.3 10.1 11.4 1.7 1.0 1.3 3.3 3.7 2.9 6.4 9.9 10.8 1.8 1.3 .8 3.5 3.5 4.0 Temperature (°F.). Absolute maximum: 3 years, 1910 to 1912 84 84 18 18 54.1 53.5 88 88 27 27 54.6 56. 4 94 87 32 32 61.5 .59. 2 106.5 91 35 36 69. 65.1 120 103 42 42 73.6 72.3 117 114 47 54 81.6 82.7 116 109 54 58 88.0 85.8 113 111 61 61 88.3 85.8 116 107 51 51 82.3 78.5 107 95 36 36 70.2 68.5 94 88 28 34 60.1 59.7 81 75 16 23 51.1 51.0 120 For 1912 114 Absolute minimum: 3 years, 1910 to i912 16 For 1912 18 Mean: 3 years, 1910 to 1912 69.5 For 1912 68.0 [Cir. 126] WORK OF YUMA EXPERIMENT FARM IN 1912. 17 Table I. — Summary of the climatological observations at the Yuma Experiment Farm 1910 to 1912, inclusive — Continued. * KlLLIXG Frosts. Last in spring. First in autumn. Years. Date. Minimum temper- ature. Date. Minimum temper- ature. Frost- free period. 1910 "F. Nov. 27 Nov. 14 Dec. 4 "F. 32 32 31 Days. 1911 Feb. 24 Mar. 31 32 32 262 247 1912 CROP CONDITIONS. The crop conditions on the project were generally favorable during the year 1912. In the early part of June the flood waters of the Colorado River caused some damage from seepage on the lands near the river, and some parts of the project were damaged by the rise of underground water from the river and from the excessively ii-rigated higher lands. The areas seriously affected by these difficulties com- prise, however, only a small part of the project and the conditions on these areas are bemg improved by the construction of a large drainage ditch. The total irrigable area of the farms on the project in 1912 was 27,592 acres, included in 470 farm units. Of this area, 13,767 acres were actually irrigated, and crops were harvested from 1 1 ,060 acres. The acreage, yields, and farm values of the crops on the project in 1912 are stated in Table II. The figures were obtained from the United States Reclamation Service. Table II. — Acreage, yields, and farm values of crops groivn on the Yuma Reclamation Project in 1912. Area. Unit of yield. Yield. Farm value. Crop. Total. -Per acre. Per unit of yield. Total. Per acre. Aver- age. Maxi- mum. Aver- age. Maxi- mum. Alfalfa hay Alfalfa seed Acres. 7,269 2,824 2,824 208 986 1,567 294 292 25 431 575 5,735 Ton Pound.. Ton ...do Bushel. . ...do ...do Ton Pound.. 27,078 814,186 3,000 416 31,. 372 55, .357 6, 602 1,252 5,800 3.73 288.00 1.06 2.00 31.80 35.30 22.50 4.30 232.00 10.00 900.00 2.00 2.50 100.00 60. 00 40.00 8.00 500.00 $10.00 .10 5.00 9.00 .75 .75 .75 5.00 .20 .$270,780.00 81,418.60 15,000.00 3,744.00 23,529.00 41,517.75 4,951.50 6,260.00 1,160.00 19,902.00 28,750.00 $37.30 28.80 5.30 18.00 23.85 26. 47 16.87 21.. iO 46. 40 46.. 30 50.00 $100.00 90 00 Alfalfa straw ^ . 10 no Barley hav.. .- 22 'ift Grain sorghurn Wheat and barley 2 Com 75.00 45.00 30 00 Cane 40 00 Cotton 100 00' Truck crops Pasture Less duplications Total 11,060 497,012.85 Average value... 44.94 1 "Alfalfa straw" is the residue left after the alfalfa soed is removed from the plants. Its farm value on the Yuma Project in 1912 was estimated at one-half that of alfalfa hay. - The yield and farm values of these grains were not segregated in the reports made by the Reclamation Service. 90341°— Cir. 126—13 3 18 CIRCULAR NO. 120, BUREAU OF PLANT INDUSTRY. CROP EXPERIMENTS. The experiments conducted at the Yuma Experunent Farm in 1912 inchided chiefly alfalfa, cotton, corn, dates, figs, and truck crops. The arrangement of the fields and the location of the crop expermients in 1912 are shown in figure 2. Cultural experiments were conducted with these and with some miscellaneous crops, and 4> YUMA EXPERIMENT EARN Tig. 2. — Diagram of the Yuma Experiment Farm, showing the arrangement of the fields and the location of the experiments in 1912. some variety testmg and breedmg^ work was done with corn, cotton, dates, and figs. Most of the experimental work was a continuation of work started in ])revious years, ami the following brief discussions of the dift'erent cro})s are based on the results obtained at the ex- periment farm, not only in 1912 but to some extent in ])revious years as well. ICir. 12tj] WORK OF YUMA EXPERIMENT FARM IN 1912. 19 ALFALFA. Alfalfa is the most important crop grown on tho project. Failure to secure a perfect stand by the usual method of broadcast seeding often enables Bermuda grass to secure a foothold and ultimately crowd out the alfalfa. The experiments conducted at the experi- ment farm in starting alfalfa have indicated that the methods de- scribed below can be expected to result successfully. It has been found that fields planted with a grass-seed drill or a gram drill with a grass-seed attachment usually produce a uniform stand. It is important that the land be first properly leveled and the grade on the fields made not to exceed 1 inch to 200 feet. After plowing, a thorough irrigation is given. Tho soil is cultivated as soon as possible after irrigation, in order to form a loose soU mulch. Spike- tooth and disk harrows are excellent implements for this purpose. Soon after the preparation of the seed bed the seed is planted suffi- ciently deep with a drill, so that all seed are deposited in the moist soil directly underneath the mulch. When alfalfa is planted with a drill, 8 to 10 pounds of first-grade seed are ample, whUe 12 to 20 pounds per acre are necessary when the seed is broadcasted. The most favorable planting season is between September 15 and October 15. When planted' at this time, alfalfa becomes well rooted before cold weather, and it is not troubled so much with weeds the following spring, as is the case with spring-seeded alfalfa. Where the soil is very sandy it is necessary to irrigate after planting in order to secure proper germination. The heaviest jdelds of alfalfa seed are frequently obtained from fields badly infested with Bermuda grass. This appears to be due to the fact that the plants produce seed better when they are far enough apart to be fully exposed to light. From these observations, as well as from results obtained from planting alfalfa in rows 24 to 30 inches apart where the plants have seeded abundantly, it has seemed desirable to try various methods of row planting for the purpose of seed production. Field plantings in rows varying from 18 to 40 inches apart have been made in order to determine the value of this method. The sandier lands on the project are relatively unproductive at first. The fh'st aim of the farmer should be to increase the fertility by the addition of silt and organic matter. Alfalfa is an exceptionally valuable crop for this purpose. It is irrigated by flooding, and con- sequently the silt in the iri'igation water is deposited uniformly over the field. The decayed roots that remain after plowing alfalfa land, as well as the surface accumulation of leaves, materially increase the humus content of the soil. The sandy areas in spotted fields are [Cir. 120] 20 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY. readily indicated by the poor growth of alfalfa. Manure or other fertilizer can be appUed to these areas until their fertihty is equal to that of the remainder of the field. BERMUDA GRASS. Bermuda grass, an introduced plant, has proved an exceptionally troublesome weed. In the Southern States, where it does not produce seed, it is considered a valuable forage grass, but in the vicinity of Yuma it produces seed abundantly. It generally covers the ditch banks, and as a consequence the seeds fall into the irrigation water, which acts as an effective distributmg agent. Bermuda grass makes an excellent pasture, but is far inferior to alfalfa for this purpose. On account of its tenacious character, the grass is difficult to eradicate when it once acquires a foothold m a field. The best method of killing it so far found has been to give a thorough shallow plowing during the summer months, followed by an occasional disking in order to cut up and clry out the furrow slice. This method has been used successfully on the experiment farm. A strong moldboard plow having a cutting edge on the land side is a useful implement for turnuig over Bermuda-grass sod. If the work is well done, a clean field can be had b}' fall. It is essential that all irrigation water be kept off the land while this treatment is being applied. If the eradication is not complete, the land can be irrigated late in the fall and a crop of small gram grown, since Bermuda grass does not grow during the winter months. If the eradication has not been complete, the process should be repeated the following summer. The cultiva- tion given to a growing crop of cotton has also been found very effective in Bermuda-grass eradication. Bermuda grass will grow where the underground water during the flood stage of the Colorado River comes practically to the surface, a condition disastrous to most other crops. It also mthstands a rather high alkali content in the soil, and on account of the rapidity with which it spreads when it once secures a scattered foothold it should prove useful in reclaiming alkali areas. On such areas, where alfalfa does not do well, Bermuda grass has a distinct value as a pas- ture crop, since the forage is highly nutritious and the plants very resistant to the eft"ects of close grazing. COTTON. For several years much emphasis has been placed on cotton work in an effort to develop varieties adapted to the clmiatic conditions of the arid Southwest. The high price which Egyptian cotton com- mands aiul the smiilarity of the conditions in the irrigated sections [Cir. 126] WORK OP YUMA EXPERIMENT FARM IN 1912. 21 of the Southwest to those of the Nile Valley led the Department to believe that Egyptian cotton could be successfully grown in the southwestern United States. Several years of breeding and selec- tion carried on at the experiment farm by the Office of Alkali and Drought Resistant Plant Investigations have resulted in the pro- duction of improved varieties of Egyptian cotton. Seeds of one variety, known as the Yuma cotton, have been distributed to farmers in southern Arizona and California. An area of 550 acres was devoted to the production of the unproved varieties of Egyptian cotton in the Imperial, Yuma, and Salt River Valleys during 1912. In addition to the breeding and extension work with Egyptian cot- ton, extensive experunents along cultural lines have been carried on at the Yuma farm by the Office of Acclimatization and Adaptation of Crop Plants and Cotton-Breeding Investigations. It has been dem- onstrated that cotton can be reproduced from mature wood cuttings and that cotton plants can be volunteered by protecting the bases of the stems with soil during the winter. A large number of hybrids between Egyptian and Upland cottons have been made and grown and different methods of makmg such hybrids have been studied. Many of these hybrids produce fiber of a very superior quality. A number of varieties of Upland cotton have been tested. During the season of 1912 two stands of roller gins were installed at the experiment farm. These were placed at the disposal of the local farmers in ginning the 1912 crop. CORN. Although corn has not yet proved a profitable crop on the irrigated lands of the Southwest, it has seemed advisable to try a number of varieties not commonly grown in the region. The yield is low and often the crop is a complete failure, due to high temperatures or dry winds at blossoming time. The corn worm is also quite injurious. A large number of varieties have been tested in cooperation with the Office of Corn Investigations, but nearly all have been discarded. In 1912, 22 varieties and selections were tested, but the results were not encouraging. Selection 157, a strain of Laguna corn, appears to be better adapted to the climatic conditions than any other variety grown at present. It has an exceptionally tough husk, which greatly reduces the worm injury. Corn yields have varied from 25 to 48 bushels per acre. DATES. The Yuma Project is included in the limited area in the south- western United States where dates can be successfully grown. Large quantities of the date seed have been distributed to settlers on the [Cir. 12tf] 22 CIRCULAR NO. 126, BUREAU OP PLANT INDUSTRY. project and considerable assistance has been given in encouraging the industry. Three thousand seedlings of desirable varieties have been planted on the farm, from which some excellent new varieties may be expected. The seedling dates include a row that was planted around the farm for ornamental purposes. Offshoots from 15 varie- ties of imported palms have been planted, and additional varieties will be planted as fast as they become available from other stations. Some difficulty has been experienced in getting transplanted offshoots to grow. The difficulty has been largely overcome by planting the offshoots (fig. 3) in soil having good drainage and then irrigating every other day during the summer months. Date offshoots seem to Fig. 3.— View showing method of rooting date-palm olTshools in nursery rows on the Yuma Experiment Farm. require that the immediately surrounding soil shall constantly be in a nearly saturated condition. Under this treatment the percentage of offshoots that do not root has been greatly decreased. FIGS. The soil and climatic conditions on the Yuma Project are well adapted to fig culture. Figs of good cpiality can be placed on the market early in the season. The Smyrna is the best fig that is mar- keted at the present time. In contrast to the common varieties of figs, which are self-pollinated, the Smyrna fig is cross-pollinated through the agency of a small insect . The requirements necessary to carry this insect through the winter and to insure its presence in |('ir. i2(i] WORK OF YUMA EXPERIMENT FARM IN 1912. 23 sufficient numbers at the right time in the spring are a serious handi- cap to the rapid extension of the Smyrna fig industry. It is highly desirable that a fig be originated that will have the excellent qualities of the Smyrna, but which, like the common varieties of figs, will set fruit without insect pollination. More than a thousand seedlings, crosses between the Smyrna and common varieties of figs, have been planted in orchard form (fig. 4) with the hope of securing some new varieties having the above-named desirable characteristics. These seedlings will begin to bear during the season of 1913. The minmium of 16° F. above zero during the winter of 1911-12 killed nearly all the seedling figs to the ground. Since equally low temperatures may occur at intervals of 10 or 15 years, it is desirable Fig. 4.— View in the fig orchard on the Yuma Experiment Farm. More than a thousand seedlings, crosses between the Smyrna and common fig varieties, are being tested. that the new wood on fig trees shall be well ripened before cold weather sets in. This was accomplished at the experiment farm during the season of 1912 by withholding imgation water from the fig orchard from about the first of August and thus preventing late growth. VEGETABLES. A garden for supplying the farm cooperative mess with fresh vege- tables has been grown each year. ^lost of the varieties of vegetables that are recommended by reliable seed houses for planting in the Southwest have been successfully grown when planted during the proper season. There is no good reason why a farmer on the Yuma Project should not have fresh vegetables from his own garden the entire year. Varieties of tomatoes, such as the Dwarf Champion, that have the fruit shaded by dense foliage, are the most satisfactory. [Cir. 12G] 24 CIRCULAR NO. 126, BUREAU OP PLANT INDUSTRY. Irish potatoes have not been extensively grown, partly on account of the frequent failures due to improper cultural methods. On the experiment farm this crop has been very successful when properly handled. It has been found that medium sandy soil is best, particu- larly if it is old alfalfa sod. The land is thoroughly irrigated early in January and plowed as soon after irrigation as possible. The potatoes are planted between Januar}^ 15 and February 1 in rows 2h or 3 feet apart, with the tubers 12 to 16 inches apart in the row. It has been found desirable to harrow the soil frequently and to withhold irri- gation water until the plants are about to bloom. One light furrow irrigation at the time of bloomiug, followed by cultivation, is usually sufficient to mature the crop. If irrigation water is kept off the field after the crop reaches maturity the tubers can be harvested as wanted for a period of six weeks. EUCALYPTUS. The protectmg levees on either side of the Colorado River are 4,000 feet apart. About 1,000 feet of this space is occupied by the river channel. The remamder of the area between the levees is subject to periodic overflows. It is desirable that as much of this land as is not occupied by the river channel shall be covered with timber. It is now covered with cottonwood and willow, neither of which is partic- ularly valuable. It has seemed probable that some of the species of eucalyptus might prove well suited to the situation and serve some- what to hold the river to its channel and at the same tune furnish useful timber. In 1910, 100 plants of eucalyptus, mcludmg the red, gray, and desert gums, were planted on this overflow land. All three of the species have done well, notwithstandmg a 3-foot overflow and ex- tremes in temperature varying from 16° to 120° F. The red and gray gums have attained a height of 24 feet and the desert gum a height of 20 feet. These trees have had to compete with the rank growth of native vegetation, consisting of cottonwood, willow, and arrow weed. There are many waste areas on the project that could be profitably planted to eucalyptus. On account of their deep-rootmg habit, these trees obtain their moisture from the underground water and are not dependent on irrigation. Every farmer would find it advanta- geous to plant an acre or two if for no other purpose than to supply fence posts and fuel. The eucalyptus trees are also desirable for ornamental purposes. A row of the desert gum has been planted around the experiment farm. [Cir. 126] WORK OF YUMA EXPERIMENT FARM IN 1912. 25 MISCELLANEOUS CROPS. In 1912 half an acre was planted to liemp in order to determine the possibilities for hempseed production on the irrigated lands of the Southwest. The soil was lacking in uniformity. On the spots of medium and heavy soils the hemp attained a height of 10 to 20 feet and seeded abundantly. The seed was harvested, but at the time of writing it had not been sufficiently cleaned to give 3deld-test weights. On account of the rapidity of growth of bamboo and the mnumer- able uses to which it can be put, several hundred plants of two large commercial species from the Orient were planted m nursery rows in 1911. These species have not proved hardy enough to withstand the low temperatures which occasionally occur. However, it is pos- sible that after the plants become well established they will be suffi- ciently self-protectmg to pass through most winters without injury. The bamboo work was extended in 1912 by a half-acre planting on low land in order to determine the adaptability of bamboo to land havmg a high underground water table. The carob {Ceratonia siliqua), a promising mtroduction from Spain, is being tested. The plants were frozen to the ground in the winter of 1911-12, but they put out branches from below the ground line and made a good growth during the past summer. PUBLICATIONS. The following publications, based wholly or in part on experiments and observations at the Yuma Experiment Farm, have been issued by the United States Department of Agriculture: McLachlan, Argyle. The brandling habits of Egyptian cotton. U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 249, 28 p., 1912. Cook, 0. F. Results of cotton experiments in 1911. U. S. Department of Agricul- ture, Bureau of Plant Industry, Circular 96, 21 p., 1912. Kearney, T. H. Fiber from different pickings of Egyptian cotton. U. S. Department of Agriculture, Bureau of Plant Industry, Circular 110, p. 37-39, 1913. Kearney, T. H. Egyptian cotton as affected by soil variations. U. S. Department of Agriculture, Bureau of Plant Industry, Circular 112, p. 17-24, 1913. Scofield, C. S. Suggestions on groAving Egyptian cotton in the Southwest. U. S. Department of Agriculture, Bureau of Plant Industry, Western Agricultural Exten- sion Circular, 10 p., 1912. As rapidly as definite results are obtained at the experiment farm, reports will be published and distributed among the farmers on the project. [Cir. 120] [Cir. 126^C] DIRECTIONS FOR COLLECTING PLANTS.' By P. L. RicKER, Assistant Botanist, Taxonomic and Range Investigations. b INTRODUCTION. Several thousand specimens of plants are sent annually by farmers, florists, nurserymen, and amateur botanists to the United States Department of Agriculture for identification. In many cases the specimens consist of small fragments of the plants, which are not adequate for satisfactory identification and are often without data. Many of these specimens if properly prepared or packed would be valuable additions to the Department collections, but as usually received they are wortliless and difficult to identify, requiring an expenditure of time entirely out of proportion to the value of the specimen or the usefulness of the information to the sender. It is with a view to the improvement of such specimens that this paper was prepared. Most of the specimens are received from persons without a knowl- edge of the methods of collecting, preparing, and packing for ship- ment, and it is chiefly to these that the directions are addressed. Abbreviated dii-ections are given on page 35 for those who have not the tune or customary outfit for the careful preparation of specimens, but more complete directions will be given first, in the hope that all correspondents will assist in building up a valuable economic collec- tion and materially reduce the amount of time and labor required to make identifications. THE EQUIPMENT. The digger. — A stout knife or garden trowel may be used for digging the plant. The most serviceable all-around tool for collecting is, how- ever, a United States Army intrenchmg tool (fig. 1),^ which is like a broad knife, the blade being 8^ inches long, 2 inches wide, and pointed at the end. It is provided with a heavy sole-leather case in which it is held by a spring and is easily attached to a belt. As a possible 1 Issued May 10, 1913. » Shown with case and hook for attaching to a belt, in the left foreground of the illustration. These can be obtained from dealers in second-hand Army equipment at a cost of about $1 each. [Cir. 126] 27 28 CIRCULAR NO. 126, BUREAU OF PLANT INDUSTRY. addition, the amateur or professional collector ^\dll find a light pick * very serviceable in a rocky soil. This is about 10 inches long and constructed similar to a mattock, but with one end pointed like a pickax and weighing only about !{ pounds in addition to the handle. If the pick is made by a blacksmith, he should be instructed to forge the shank into which the handle is wedged not less than H inches in length. The collecting case. — For those who are collecting j)lants in the field two methods are available, a third method, that of carrying a press and driers and putting the plants in as collected, being discarded by most of the experienced collectors. First, for short trips on which it Fig. L— a collecting portfolio, bundle of plants in press, plant digger and case, and folding stove used in drying plants, with collapsible smokestack In canvas carrying cases. is expected to obtain only a few plants a collecting box may be used. This is made of tin and is cylindroid in form, with a cover hinged on the side, and is carried by a strap attached at the ends. For more extended trips or one on which it is expected to obtain a large number of specimens some foiTQ of collecting portfolio is pref- erable. (Fig. 2.) A very satisfactory one may be constructed from two pieces of heavy binder's board or leather board, measurmg 12 by 17 inches. To the long edge (the back) of one, near each end, should be riveted a buckle, and to the correspondmg portion of the other board should be riveted two straps about 6 inches long. Thus, the two parts when buckled together have a back adjustable to the num- ber of specimens obtained. To the middle of the upper side of one ' These tan he obtained from hotanical supply companies at a cost of about $1.75 each. ICir. lliUJ DIEECTIONS FOR COLLECTING PLANTS. 29 cover should be riveted another strap, which may be put through a buckle fastened to the corresponding point on the other cover, to keep the covers together. A much more convenient fastening for the strap, however, is a button, similar to that by which the curtains are fastened on a carriage. Short slits to go over the button should be cut in the straps at intervals, doing away with the inconvenience and waste of time involved in fastening and unfastening a buckle. A few inches on either side of the middle strap and its fastening should be fastened the ends of a leather handle. Such a handle should be attached to each cover, so that the two can be used in temporarily holding the covers together. Thus, one avoids the necessity of bother- ing with the strap every time a plant is put in the portfolio. The portfolio covers should be sufficiently thick to avoid much warping at the corners. The warping can be largely prevented by riveting a thin strip of iron or steel about three-fourths of an inch wide to the inside Fig. 2. — A collectinfj portfolio, showing plan of construction. of the upper and lower edges of the portfolio covers. If the covers are of heavy binder's board, the total weight of the portfolio is con- siderable, but this may be replaced by tough fiber board, such as is used in light extension cases. The boards may be covered with enamel cloth or leather, with flaps for the protection of the specimens in rainy weather. The side flaps should be 10 inches wide and the two end ones 7 inches wide, the latter being held together by a strap. The portfolio should then be filled with thin papers, which when folded should measure 11 J by 16| inches. Newspaper stock is best for tliis; in fact, ordinary newspapers cut to this size may be used with advantage. The above sizes should be maintained as a maxi- mum in order that the resulting specimens shall not be larger than the standard mounting sheet, which is 11 1 by 16^ inches. The unused sheets can be kept in place and easily slid from under the flaps on one side of the portfoho as needetl, the flaps on the opposite side [Cir. 126] 30 CIRCULAR NO. ^F 126, BUREAU OF PLANT INDUSTRY, aiding in holding the filled sheets in position. The strap holding the filled sheets need not be fastened each time a folder is placed under them, as in carrying the portfoHo short distances the handles of the portfolio give sufficient pressure to hold the specimens in place. Two portfolios may be carried if many specimens are to be obtained. They will hold all that one person can collect in a day. The press. — The best form of press (fig. 3) consists of two slat fi-ames, 12 by 17 inches, made from pieces of ash three-sixteenths of an inch thick and three-cjuarters of an inch wide, the slats to be i)laced about an inch apart, the ends and all intersections being fastened with Fig. 3. — A plant press before strapping together, showing construction of the slat frames and the sheets of corrugated cardboard placed between the pairs of driers which separate the folded thin sheets containing the plants. three or four wire brads, which should be securely clinched. Small rivets may be advantageously substituted for the wire brads in the ends of the strips of wood, since with hard usage the wdre brads are more Hkely to pull out. The driers should be of heavy gray blotting paper or felt carpet paper, cut llj by 16| inches. Pieces of corru- gated strawboard (fig. 3), preferably with both surfaces smooth, cut to llf by 16| inches, mth the corrugations running longthA\ise, placed between the blotters will be found to be of inestimable assist- ance in drying the plants. Two ordinary leather trunk straps may be used to secure the necessary pressure. Recent experience has, however, demonstrated that a heavy canvas trunk strap having a LCir. 126] DIEECTIONS FOR COLLECTING PLANTS. 31 buckle with a fiat tongue and corrugated edge is more satisfactory than a knither strap. SoUd board presses have been used, but they are more hkely to cause a sweating and blackening of the plants on the outside of the bundle. SPECIAL DIRECTIONS FOR PREPARING AND DRYING PLANTS. Herhaceous plants. — All herbaceous specimens (i. e., those without a woody stem) that are not over 3^ feet tall should be collected entire, including root, stem, leaves, flowers, and fruit when available. j=?>^ Fig. 4.— a specimen placed between thin sheets, showing tlie metliod of folding a tall plant before pressing. Wlien the stem of a plant is too long to go into the sheets it should be folded into 12 or 15 inch lengths, beginning at the root end to fold (fig. 4). The folds may be held together by strips of thick paper about 1 by 3 inches, with a slit in the middle, the folded ends being inserted through the slit. When herbaceous specimens are over 3J feet tall it is rarely necessary to collect the whole plant. The height of the plant should be noted on the folder or in a notebook and placed on the label of the plant when it is mounted . Many tall specimens are too thick at the base to be collected entu"e. In such cases a piece nil-. 12*;] 32 CIKCULAR NO. 126, BUREAU OF PLANT INDUSTRY. of the top of the plant about 15 inches long should be collected, together with leaves from the mitldle or base of the stem, showing the maximum size and any variation in shape. If the root is bulbous or thick and fleshy it should be split in the middle, and if necessary to further reduce the thickness (which preferably should not be over one-fourth of an inch) the inner portion may be scooped out with a knife. All dirt should be removed from the roots and the specimen should be laid out smoothly in one of the foldei-s and placed under the flaps on the side of the portfolio reserved for this purpose. Woody plants. — A branch from a woody specimen bearing leaves and flowers, or fruit when obtainable, should be about 12 to 15 inches long. Sterile shoots and suckers bearing leaves should be collected when the leaves differ from those of the flowering branches. Aquatic plants. — Some aquatic plants will stick to the thin drying sheets so that they can not be removed without injuring the speci- men. These should be placed in clean water and floated out on a piece of clean white paper that can finally be attached to the mount- ing sheet, placed in one of the thin driers, and the plant covered with a piece of the finest mesh white cheesecloth or unbleached cotton cloth. When dry this can with care be peeled off without injuring the specimen. Succulent plants. — Some plants of the orpine family are very diffi- cult to dry perfectl}^ under ordinary conditions, and after they are apparently dried they have been known during damp weather to send out shoots while in the herbarium. This may be prevented by immersing the plant for a few seconds in boUing water to kill the tissues before it is put in the press. Fleshy plants of the cactus family may be treated in the same way if the joints are to be pre- served in boxes. Excellent specimens are also made by splitting the joints laterally, scraping out the inside, and pressing the halves. Seed. — Small quantities of seed may be collected m strong manfla envelopes. For larger quantities, bags made from cheesecloth or cheap cotton cloth may be used. It is very desirable in collectmg seed that a herbarium specunen from the same plant as the seed be collected to assist in a correct identification, the seed and specimens being given the same number as soon as collected. If possible the specimen should be preserved in some herbarium where it can be consulted should any question arise as to the identification of the seed. Pressing and drying. — Upon returning from a collectmg trip, the plants in the folded sheets should be put in press as soon as possible. They will, however, i-emain in very good condition in the portfolio as long as 12 or 18 hours. Two driers should be placed between the [Cir. 126] \"^ ',■■ DIRECTIONS FOR COLLECTING PLANTS. 33 folders containing the plants, and, if the plant is very fleshy, three or four driers may be used to advantage. Wlren the corrugated pieces of strawboard previously referred to are used, only one drier should be placed between it and the folder containing the specbnen (fig. 3). The straps should then be adjusted around each end of the press and pulled up as tight as practi- cable without crush- ing the plants. Sub- sequent These may be obtained from dealers in campers' supplies. H'ir. 12tj] DIEECTIONS FOE COLLECTING PLANTS. 35 The locality should be so definite that the spot can be revisited if necessary. A plan of the mimediately surrounding region may advantageously be made in the notebook to aid in relocating indi- vidual plants. Notes should be made of the habit and size of the plant, particularly when the whole plant is not collected. The data should be written on the sheet in which the plants are pressed or on a temporary label pro"sdded for the purpose, or, if more convenient, the sheets may be numbered and the data all entered in a notebook at the tune the specimens are collected. Specimens for identification should be prepared in duplicate, both given the same number, and one sent with full data, the determinations being returned by num- ber. Great care should be taken that all specimens bearmg the same number are specifically identical or are from the same plant, as much confusion has been caused by collectors distributing two or more species under the same number. For convenience in citation by publishing botanists, those making herbaria of their own or who are collecting plants for distribution should number all their collec- tions consecutively from 1 upward and should not begin at 1 each year, as tliis often leads to confusion. A field notebook should also be kept, in which is a list of the numbers, followed by the names when the plants have been identified, the date and place of collecting, the altitude when from mountains, and the habitat and kind of soil in which they grew. Cultivated specmiens should be accompanied by data as to the native country or trade fu'm from which the speci- mens or seeds were originally received. It is also important that any local or trade name should be given when known, together wdth any economic uses of the plant. A plain label about 1| by 4 inches, bear- ing such data as necessary, is preferable, with the region in which they were collected printed at the top, as "Plants of Arizona." If many labels are to be written, other data may be printed to save time. SHIPPING SPECIMENS. Pressed and dried specimens should be shipped in the thin sheets in which they were collected, and the bundle of sheets placed between pieces of binder's board or heavy cardboard and securely wrapped and tied. Where it is impracticable to make a good herbarium specmien of the plant, the specunen should be carefully wrapped in damp sphagnum moss and inclosed in oiled or parafiin paper. If these materials are not available, moistened newspapers may be used for wrapping the plants and the package then inclosed in stout wrapping paper. Packages to be sent by parcel post must not exceed 11 pounds in weight or 72 inches in length and girth combined. [Cir. 126] ADDITIONAL COPIES of this publication -tS. may be procured from tiie Superintend- ent OF Documents, Government Printing Office, Wasliington, D. C, at 10 cents per copy U. S. DEPARTMENT OF AGRICULTURE. BUREAU OF PLANT INDUSTRY— Circular No. 127. WILLIAM A. TAYLOR, Chief of Bureau. MISCELLANEOUS PAPERS. The Work of the Delta Experiment Farm in 1012 . Silver Scurf, a Disease of the Potato The Dasheen, a Root Crop for the Southern States JOHN P. IRISH, JR. I. E. MELHUS ROBERT A. YOUNG Issued May 17, igi3. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1913. BUREAU OF PLANT INDUSTRY. Chief of Bureau, William A. TA-i-LOR. Assista7it Chief of Bureau, L. C. Corbett. Editor, J. E. Rockwell. C/i/f/ C/(rfc, James E. Jones. \ [Cir. 1271 2 [Cir. 127— A) THE WORK OF THE DELTA EXPERIMENT FARM IN 1012/ By John P. Irish, Jr., Farvi Superintendent, Office of Western Irrigation Agriculture. INTRODUCTION. DESCRIPTION OF THE REGION. The Delta Experiment Farm is located on the Lower Jones Tract, one of the islantls in the south-central portion of the San Joaquin Delta. The tract is typical of the general San Joaquin Delta lands. The soil is a very light peat, characteristic of most of the reclaimed land on the San Joaquin, side of the delta. These lands lie at sea level and before their reclamation were flooded during the major portion of the year. The delta of the com- bined Sacramento and San Joaquin rivers occupies an aggi-egate area of about 250,000 acres. Reclamation is accomplished by leveeing to exclude tidal and flood waters and by the installation of ditches and pumps discharging into the river to furnish drainage and control the ground-water level. The drainage system consists of main (h-ainage ditches or canals from 18 to 30 feet wide and from 7 to 12 feet deep, into wliich empties a secondary system of hand-dug ditches 3 to 4 feet wide and 4 to 6 feet deep. These secondary ditches divide the land into blocks or fields of from 10 to 100 acres, the average unit being 50 acres. Since the rainfall of the section averages only about 15 inches per year, irrigation is necessary for the production of crops. The sec- ondary ditch system is used both for drainage and as head ditching for irrigation, the water being drawn into it from the river channels by means of siphons over or headgates through the levees. The climate is mild, and plant gi-owth is continuous throughout practically the entire year. This somewhat complicates the problem of controlling weeds and plant diseases, the latter being specially difficult to eradicate because of the continuous gi-owth of many of the host plants. 1 Issued May 17, 1913. The Delta Experiment Farm was established in 1011 at Middle River, 14 miles west of Stockton, Cal., on a tract ofland comprising about 100 acres provided by the Delta Association of California. Thisassociation is composed of farmers who are operating in the delta region, and the organization cooperated with the Bureau of Plant Industry in the work here reported. The University of California is also conducting inves- tigations on the same tract, about 50 acres being used for this purpose. Actual operations were begun in the spring of 1912. The work in which the Bureau of Plant Industry was directly concerned in 1912 was devoted to experiments with potatoes. This work was under the immediate supervision of Mr. John P. Irish, jr., who was detailed from the Bureau of Plant Industry. [Cir. 127] 3 4 CIKCULAR NO. 127, BUREAU OF PLANT INDUSTRY. CROPPING SYSTEMS. The principal crops gi-own are potatoes, barley, beans, and onions. A large acreage is also devoted to asparagus, and celery is gi-owii to some extent. The principal cash crop of the region is potatoes, 52,000 acres being devoted to this crop in the delta region in 1912. The barley acreage, while large, usually gives a much lower acreage return than potatoes. The present cropping system on the San Joaquin side of the delta is based principally on a 2-year rotation of potatoes and barley. In many cases tenants holding term leases grow barley hay, beans, and onions instead of barley in alternation with potatoes, but in a majority of cases the barley-potato rotation is the accepted one, the effort usually being to devote as much land to potatoes as possible, although the exigencies of the potato market and the decrease in potato yields sometimes cause individual owners to hold their lands in, barley for more than one season in an effort to improve conditions. The lands are farmed largely on an annual tenancy system, the potato tenant moving from farm to farm with the rotation of the crops, while the barley tenant farms each year an aggregate area comprising five or more potato farms. The unit area of potato farm- ing is from 100 to 300 acres. Potatoes are planted from the latter part of March till early June, the bulk of the planting being done in April and May. The harvest ordinarily extends from the last of September to the first of January, the potatoes being dug as the market requires them. The land is given a preliminary plowing in the early spring to check the weeds and improve the soil tilth and is then allowed to lie from one to three months. Planting is done while plowing, the seed being dropped by hand in the alternate furrows. The seed piece is commonly cut to two eyes. The spacing is from 30 to 32 inches between the rows and 8 to 12 inches between hills, giving about 20,500 plants to the acre as a maximum stand. The ground is harrowed from one to three times before the plants are large enough to be damaged. When the plants are large enough to defhie the rows, ditches 10 inches wide with an average depth of 18 inches are cut between two rows 60 to 70 feet apart. Water is later run into these ditches from the head ditches, regulated to fill them nearly full, and irrigation is accom})lished by lateral percolation. Three to five irrigations are given during the gi-owing season, the number depending on the soil and climatic conditions and the h(nght of the ground-water table. The crop is given one or two hand hoeings, one to three cultivations, and one hilling. Ilarvesthig is done by hand with ])otato hooks, the sorting being done during the digging, the culls usually being left in the field. This method, while tedious, LCir. 127 j WORK OF THE DELTA EXPERIMENT FARM IN 1912. 5 expensive, and rather wasteful, is made necessary by the physical peculiarities of the peat soil, in which no machine digger so far tried has done effective work. PREVIOUS INVESTIGATIONS. Progressive decrease in potato yields and a lowering of quality under the present methods of cropping led to an investigation of the potato situation in 1908 by the Office of Cotton and Truck Disease and Sugar-Plant Investigations of the Bureau of Plant Industry, and the results were published in 1909.^ These results seemed to indicate that the decrease in yield and quality was due largely to pathological conditions. Some preventive measures were described, but it was stated that further investigation was necessary, particu- larly in comiection with the improvement of cropping systems and cultural methods. EXPERIMENTS WITH POTATOES IN 1912. In the spring of 1912 work was begun by the Office of Western Irri- gation Agriculture looking to the improvement of cropping methods on these lands. Since potatoes are the crop upon which the land- holders depend for maximum returns, their place and duration in any rotation are of great importance. In the work for the crop season of 1912 the main aim was to determine what method of tillage, ferti- Hzation, and rotation may be expected to give the best results in potato production and weed control. It was thought that the potato wilt might be controlled by increas- ing the vigor of the plants and that the damage from scab might be reduced by acidulating the soil. For tliis reason the experiments included methods of stimulating plant growth, particularly through the use of commercial fertiUzers, and also included methods of acidu- lation. The Burbank variety was used in aU the trials. The land used in 1912 comprised about 32 acres in three fields (fig. 1), as follows: Field F, 15.5, field G, 6.6, and field I, 10 acres. This land had been planted to potatoes in 1911 and was very weedy when taken over for experimental purposes. In most of the experiments determinations were made of the stand, the extent of potato ^\dlt {Fusarium oxysporum), the extent of scab, a disease due to a fungus {Oospora scabies), and the yield of marketable tubers and of tubers of unmarketable sizes. The method followed in these determinations and the meaning of the figures used in the following tables are described later. I Orton, W. A. Potato Diseases in San Joaquin County, Cal. U. S. Department of Agriculture, Bureau of Plant Industry, Circular 23, 14 p., 1909. [Cir. 127] CIRCULAE NO. 127, BUREAU OP PLANT INDUSTRY. During the growing season a count was made of the actual stand on the plats, and during the latter part of September a count was made of the wilted plants. In tabulating the results of these counts the percentage of stand was determined by computing the percentage OF m ENT5 C( )r in n IC 'M ^L DE f EX :pth ^LOWI ^ERir G F D n ■|L .\: :ti R A d ) G \i E M n Aij JU Rl N i- — T 1- r*- rr T L J 1^ HAY U r OF ■ F 10 V\ING 1 1 D FL F:ir in 1E^ NG yd Fig. 1.— Diagram showing llie fields used in llie experiments with ^potatoes |at the Delta Experiment Farm in 1912. . of a perfect stand, estimated by the distance between rows and hills on each plat; the potatoes having boon planted at the usual dis- tances as described above, a stand of 20,500 })lants per acre was taken as 100 per cent. The ])ercentage of wilt was computed as the ratio between the wilted plants and the actual stand. [Cir. V27] WOEK OF THE DELTA EXPERIMENT FAEM IN 1912. 7 In October and November a sample digging was made of four rows equidistant across each plat. In tliis digging aU the potatoes in each liill were gathered, regardless of size or condition, and a sorting was made in wliich all potatoes of marketable size and in marketable condition were segregated, weighed, and classed as marketable. The cuUs were then sorted and weighed, being classed as undersized or scabby. The undersized scabby potatoes were classed as undersized, and only the potatoes wliich but for the scab would have been mar- ketable were classed as scabby. In computing the results of the sample digging, the percentage of scab and unmarketable sizes was determined in terms of the marketable potatoes obtained in tliis digging, and the results were taken as being representative of the plat as a whole. In December and January the final digging of marketable potatoes was made, and this yield when added to the marketable potatoes obtained during the sample digging and computed to an acreage yield is carried in the tables as the market yield. TESTS OF TILLAGE METHODS WITH GREEN MANURE AND COMMERCIAL FERTILIZERS. Alternate lands in field F were plowed 6 inches and 12 inches deep, respectively, m February. The north half of the alternate pans was sown to vetch early in March, and the south half of the same pair was sown to barley. Certain blocks of these lands were treated with commercial fertilizers. The vetch made very little growth, but the barley reached a height of 15 to 18 inches by the middle of May, when the lands were replowed. In addition to the fertilizers applied in March on certain of the green-manured and bare plats, fertilizer was applied in the rows of potatoes on certain other of the green-manured and bare plats at planting time, early in June. The fertilizers used in both cases were K2SO4 (potassium sulphate), KCl (potassium chlorid), and CaHP20e (superphosphate). All were very high grade. All planting on field F was at a depth of 6 inches. The results of this test are given in Table I, together with comparisons of the results obtained with the same green manures and tillage methods but without the use of com- mercial fertilizers. For convenience in comparing averages, the data are given in con- densed form at the close of the table where the averages of aU the results obtained with commercial fertilizers are compared with the results secured on the unfertilized plats. [Cir. 127] 8 CIRCULAR NO. 127, BUREAU OF PLANT INDUSTRY. ) o »o s "•^oT '^ a o M p. - o a t^ o (N lo r^ rr to i^ lo -^ 11 "3 ^ O 0>0 CM -^ ■d 2 ^ o ^ ^ -S A O 9f C3 cct- ON _o P. ft PQ ^ >0 O CT> "J* »-( • -CO . . S3 .-a o 0| Oi O: Oi t— ( lO s ^ H t, o p o t- 2 CO »o 1-^ Ol »0 fO lO t^ Ci ic CO lo t^ as »o 1 "35 i-H-r)^:D C^iOrHTt^O CNiO.-4'l-CO C^wi 1 txtu o cor- - (N cot^ CO • -M ^§p-" coi- -IN OS "O .. w n o . . t--OOGiO O(Mi0»00C 1^ C^J '.O ^ X Oi !N t- J3 CMlO ^lO C^iO-O) cs^ w a 03 ■* Th CO d ^ J5 O 2 a 0/ CO a o . C<1 -H cs w .- ■* t.1 O o t-^ CO r* o 1 1— I 1 PI t^ & 'd a o Oi CO O ^ CO CO CO OC . .CO . . . . cs O O -HCO — iO -< .g "? tH GC' »0 CO CM »C T-4 1^ lO (-^ ira oi ci c ■* 00 J3 03 ^-,. - IN CO "O " .-H Ol I- -. rH o3 ft 1 •s g2 . .CO . . . . -^Ji 'S* iO N r-1 C^ (M C^ CN ** OS t^ o»c . . . . oo lO -^ ocsco c4 «j CO t^ ^CS COOO^rH COi^ Of a C3 •^ «5 00 M S Qi P< 6J5 P. ^, OOiOCC 003IM OOO -ho: rfCOiC oo ^ C3 03 Q^ .... CO tS V.I ^ -^ *Jt "i CO O CO !> OC CO O CO OC ~ >-i cooo c4oo « « s CO X' - (N .-H 00 l^ K I— t H 1 _o m ce I^ t- r^ ft a £ c a C5 03 ooc »0 (M O O lO «0 1^ O O (M c>» ft O^ 03 o> a *^ ■ to .^ ■ • ■ u3 ■+-'■•• 3 ■^ . s°i Sod 6V, S d d d'2 gd ■o ■^ 3 ^" ■a ■ ''**. ■ o^ * o^ * ■2 f ■o * ■c . a I.S ij'"? 2 'elV a i;'d 1— I • 3 5« ^ O C3 w > .P g; o -P • CS^-O^ ^ jij-aJ^ oi/i-a^ 1 I- H CD • ^".-. 03^ O — a-^ "da- t— 03 ;- 03 a S :" 03 a 03;- c; w?:S^ m;SS U> M<-SU> ■s§ i .S rt 2 C3 ftfe Q> 'C si e3 o ft JS S "3 «M*o ft o w §1 M O g a a J3 3 3 fco c do's o o d o's; o o o o feQG cagGQ PQSSCfi CO ', -r ft *^ *^ •-I 3 o o 1 a 1 Ph Ai 1 — ft It ft to 03 .2 ft oo 03 ft p o o < S o p < o !5 P O • t— 1 • • CO 1 CO* CM C^ §5 ■ 1^ ■ CO CM CO* X *o CM 3- CO O r-l O -^ CM 00 lO O x" '^ X CM 1-^ CM CO -^ 1- - - CO ^ :0 N"i -^ i^ CO o o; CM CO , . . . lOt^ CD ■«»■ .— 1 iC '^ t-- t-- CM CO "0 1-- -. - CO "^ lO cr, w 19.1 31.0 58.0 70.0 6,445 28.7 ujtO t- lO ^ O CM O I^ ■ . . . CM CO »c oi CT) o; ^ Ol CM CM CM CO l^ - - CO X 1- >0-H 00 lO I-OCOOCOCM XiO.-it- oiooOTpClCM I-^OcOiO CMCOCOI- -- T-iCM COCO COt^COOr-HCM OOrHl^ . . . . I- r- .... CM-^COiOCMI-^ COiO'^'^ cocot^r^ -- i-tCM t-iO I-OCMOr-iO COCMCl-^ . . . . O "^ .... COCOX1-- ^ " .-tCM I^CO o »o o o lo a> 1-H -^ CO f-i 00 oi ^ Ol CO Cl t-^i-HCOiO COCMXt^ -.-s .-hCM a 1- c e a: \Marketable yield per acre, / pounds. Unmarketable: Undersized per cent . . 1 Scabby do.... > O ■ SI X. c a J c , ^ a — & - is? 1 I fir. IL'TI WORK OF THE DELTA EXPERIMENT FARM IN 1912. 9 The results of various methods of tillage and green manuring and the same methods with applications of potassium sulphate, potassium chlorid, and superphosphate show some significant differences. The most notable of these is the rather uniformly higher yields with superphosphate under all conditions of tillage and the uniformly better yields in proportion to stand wliich were obtamed by applying the fertilizer in March over applying fertilizers at planting time. Deep plowing showed better results than shallow, and green manure gave generally better yields than were obtained from the bare fields. THE USE OF VARYING QUANTITIES OF SUPERPHOSPHATE. An expermient was conducted to determine the effect of applymg superphosphate at different rates on land to wliich various tillage methods were applied. Each of the eight plats used in the experi- ment contained one-tenth of an acre Table II. The results are given m Table II. — Jiesults obtained with potatoes on land treated with varying qnantities of superphosphate at the Delta Experiment Farm in 1912. Treatnienl. Plowed 12 inches deep. Plowed 6 inches deep. Green manure. Bare. Green niannre. Bare. Superphosphate per acre pounds . . Stand per cent . . Wilt do.... Marketable yield per acre pounds . . Unmarketable: Undersized per cent. . Scabbv do 600 .34.7 97.0 8,805 21.9 8.0 1,200 34.0 81.0 7,105 15.3 2.4 600 31.7 75.0 10,402 12.9 2.6 1,200 38.7 50.0 9,030 20.9 4.5 400 44.9 84.0 8,575 23.6 11.8 800 36. 3 77.0 8,105 18.6 9.9 400 36. 3 77.0 9,360 21.7 10.3 800 33. 6 SCO 8,555 39.8 11.1 It is seen from Table II that varying the rates of superphosphate application resulted in no consistent differences in yield. The growth of haulm, however, was noticeably increased by the heavier applica- tion. ACIDULATION OF IRRIGATION WATER. ^ With a view to studymg the effect of acidulatmg the u-rigation water on the growth of potatoes and more particularly on the action of the scab organism, Oospora scabies, an experiment was conducted on four tenth-acre plats in field F. Commercial sulphuric acid was used. The method used consisted in the placmg of a plank 10 inches wide and 6 feet long in the ditch at the head of each plat, the 1 This work was planned in accordance with suggest ions of Mr. Karl F. Kellerman, Physiologist in Charge of the OfTice of Soil-Bacteriology and Plant-Nutritiou Investigation. 91517°— Cir. 127—13 2 10 CIRCULAR NO. 127. BUREAU OF PLANT INDUSTRY. plank being pegged Avitli the upper end 2 inches under water and mth'a fall of 4 inches to its length. Carboys containing a weighed quantity of acid were placed at the head of each plank with stoppers arranged to drop the acid at a rate designed to empty the carboys in six hours. A green-manure crop was plowed under on all the plats m early spring. The acid applications were made at the time when the tubers were beginning to fonn. The results are given in Table III. Table III. — Results of aridulation of irrigation water on potatoes at the Ddta Experi- ment Farm in 19U. Rate of acid application per acre pounds . Stand per cent . Wilt do... Marketable yield per acre pounds. Unmarketable: Undersized per cent. Scabby do . . . Depth of plowing. 12 inches. 6 inches. 12inches. flinches, 500 32.3 58.0 5,212 13.4 10.6 500 31.4 69.0 5,992 24.2 900 30.9 64.0 4,247 IS. 5 9.4 900 36.8 70.0 5,622 31.8 4.4 While the results of acidulation show no increase in yield or reduction in scab, they give other indications which are of interest. These applications were made when the tubers were beginning to form, so that the percentage of stand has no relation to the applica- tion of acid. The immediate effect of the acid was a marked increase in the vegetative vigor of the plants. This stimulation continued during the remainder of the growing period, so that at harvest time the haulm was double that of any other plats in the field. The wilt infection on these plats was not only uniformly lower than on the rest of the field, but the growth of the infected plants was much less depressed. When the infected haulms on the rest of the field were dead, the infected plants on the acid-treated plats w^ere still alive and with a large portion of their total leaf area green. These results, taken in conjunction with the lateness of the application, the moderate \)v\ce of sulphuric acid, and the ease of application, suggest a prom- ising field for further work. DEPTH-OF-PLOWING TEST. An experiment to test the effect of ])l()wiiig at varying (lc})ths was conducted on field G. The plowing was done in March, the dei)ths being 6, 12, and 18 inches. The potatoes were planted in June at a uniform depth of G inches. Each part of the experiment was dui)li- cated on quarter-acre plats, so that each result given in Table IV is the average of two plats. [Cir. 121] WORK OF THE DELTA EXPERIMENT FARM IN 1912. 11 Table IV. — Results obtained with potatoes at three different depths of plowing at the Delta Experiment Farm in 1912. Depth of plowing. 6 Inches. 12 inches. 18 inches. Stand per cent. . 30.6 4,370 22.5 5.3 28.6 5,405 17.7 7.2 29.4 Marketable yield per acre Unmarketable: pounds.. percent.. 5,978 16.7 Scabby ..do.... 5.6 The yields, as shown in Table IV, increased with the depth of plowing, but the difference between th el 2 and 18 inch plowing was probably not enough from the standpoint of one season's yield alone to justify the added expense of the IS-inch plowing. DEPTH-OF-PLANTING TEST. On field I an experiment in depth of planting in connection with two depths of plowing was tried. Alternate lands were plowed, respectively, 6 and 12 inches deep in March, and potatoes were planted in June, 4 and 8 inches deep on each kind of plowing. Eight half-acre plats were used in this test. The yields of marketable tubers per acre are shown in Table V. Table V. — Yields of warketahle potatoes obtained from planting at different depths at the Delta Experiment Farm in 1912. Treatment. Plowed 12 inches deep. Plowed 6 inches deep. Green manure. Bare. Green manure. Bare. Depth of planting inches. . Yield per acre pounds. . 4 6,822 8 5,610 4 7,055 8 7,096 4 3,264 8 4,091 4 5,502 8 7, 057 It is seen that with shallow plowing deep planting gave better results than shallow planting and that the results on the deep-plowed plats were contradictory. COMPARISON OF LOCAL AND IMPORTED SEED. Conditions for storing seed potatoes in the delta region are ordi- narily not favorable. As there are no facilities for cold storage or for storing in pits, seed carried over from harvest to planting time is frequently weakened by sprouting. In order to determine the effects produced by using locally grown seed as compared with those result- ing from the use of seed of the same variety shipped in from Oregon ICir. VH\ 12 CIRCULAR XO. 127, BUREAU OF PLANT INDUSTRY. at planting time, an experiment with the two classes of seed was conducted on 22 tenth-acre plats. The seed imported from Oregon had been stored there in pits during the winter of 1911-12. Four plats of this seed were planted, two on land plowed 6 inches deep and two on 12-inch plowing. In the same series, nine plats plowed 6 inches deep and nine plats plowed 12 inches deep were planted to locally grown seed. The results are given in Table VI. Table VI. — Results obtained with potatoes grown from Oregon seed and from locally grown seed at the Delta Experiment Farm in 1912. Stand per cent. WUt do . . . Marketable yield per acre pounds . Unmarketable: Undersized per cent . Scabbv do. . . Oregon seed . Deep plowed. 39.1 S3. 10,204 37.6 .5. 1 Shallow plowed. 37.4 81.0 9,440 1.5.6 ti. 6 Locally grown seed. Deep plowed. 31.0 70.0 6,445 24.2 6.4 Shallow plowed. 30.0 74.0 6,272 20.5 .5.7 Table VI shows the differences in yield between pit-stored Oregon seed, sound and unsprouted, shipjjed in at planting tune, and local seed weakened by si)routing. The yield from Oregon seed was decid- edly higher and the stand was better. It is also noticeable that deep plowing again gave better results than shallow plowing. SUMMARY. The work conducted on potatoes at the Delta Expermient Farm in 1912 resulted in several important indications. Some of the more prominent facts noted are the following: (1) Of the three commercial fertilizers used, superphosphate was the only one which markedly increased the average yields as com- pared with those obtained from the unfertilized plats. (2) Better average results were obtained from the use of commer- cial fertilizers when the application was made in March than when the fertilizer was applied at planting time, in June. (3) Plowing under a green-manure crop of barley resulted in in- creased average yields. (4) The yields were generally better on 12-inch plowing than on 6-inch plowing. (5) No consistent yield differences were obtained where super- phosphate was applied in varying quantities. The growth of haulm, however, was increased by the heavier applications. (6) Acidulation of the irrigation water with sulphuric acid resulted in no apj^reciable reduction in the action of potato scab and no in- ICir. 121] WORK OF THE DELTA EXPERIMENT FARM IN 1912. 13 crease in yield, but the vegetative vigor of the plants and their resist- ance to wih were markedly increased. (7) In the test of 6, 12, and 18 mch plowing the yield increased with the depth, but the difference between the 12-inch and the 18-inch plowing was not sufficient to justify the added expense of the latter. (8) On the shallow plowed land 8-inch planting gave higher yields than 4-incli planting, but no consistent difference in yield was noted on the deep-plowed land. (9) Decidedly higher yields were obtained from sound Oregon seed than from locally grown seed, the difference appearing to be due to the deterioration of the locally grown seed resultmg from miproper storage. (10) With the exception of the plats treated with sulphuric acid no consistent differences were shown in the wilt percentages, nor were any consistent differences shown in the scab percentages. (11) One significant feature of the season's work was the uniformly thin stand on the fields. An examination of the plats following plant- ing tune showed that the decrease in stand on the old land was due to the very rapid rotting of the seed in the ground. The mvestigations along this line durmg the crop season of 1912 lead to the conclusion that the rotting of the seed pieces, resulting in thin stands and weak plants, is perhaps the most important factor causing decrease of yield. [Civ. 127] [Cir. 127— n] SILVER SCURF, A DISEASE OF THE POTATO.^ By I.E. Melhus, Pathologist, Cotton and Truck Disease and Sugar-Plant Investigations. INTRODUCTION. It seems desii'able to call the attention of potato growers to a disease of the potato which, though little known, is becoming widespread in the United States. This is the silver scurf, a tuber trouble caused by a species of SpondylocUidium which attacks and destroys tlie peri- derm, causing discoloration and loss of moisture. According to infor- mation in hand it seems to have secured a foothold throughout the Eastern States without its presence being generally recognized. Introduced in all probability from Europe, it has spread rapidly and will have to be reckoned with as one of the factors in the deterioration of this crop. While the inconspicuous appearance of silver scurf has permitted it to be overlooked, it can readily be recognized after a little experience, and everyone interested should learn to distinguish this disease from the several others which occur on potato tubers. APPEARANCE OF DISEASED TUBERS. Silver scurf in its earlv stages of infection under moist conditions causes blackish olive patches on the surface of the potato, which may, and often do, cover considerable areas. This more or less superficial growth is the brown-mold stage and consists of the conidioj^hores and conidia of the fungus. This stage is well shown in figure 1. In the later stages of the development of the disease small black sclerotia are formed on the infected areas instead of the superficial growth of conidiophores. The surface of the discolored areas when clean takes on a silvery or glistening appearance, which is due to the raising of the cells of the outer layers by the mycelium of the fungus. This disfiguration of the surface is later followed by depressed patches of greater or less extent. The early stages in the formation of such spots are shouii in figure 2, while the late stages are evident in figure 3. As the disease progi'esses the infected areas increase in diameter and the fungus penetrates deeper into the tuber. As a result it is not uncommon to find the whole surface of a tuber discolored, shrunken, and shriveled like the lower tuber in figure 3, while the outer layers of the periderm are loose and sloughing off. 'Issued May 17, 1913. [Cir. 11'7] 15 16 CIRCULAR NO. 127, BUREAU OF PLANT INDUSTRY. DESCRIPTION OF THE FUNGUS. The fungus causing silver scurf has two more or less markedly dis- tinct stages in its development: The brown-mold stage (fig. 1, o, b, and c), which appears first and consists of the conidiophores and conidia, and the sclerotial stage, wdiich follows later and consists of numerous small black sclerotia biiiied in the outer cells of the dis- FiG. 1. — Tho brown mold stage nf Spondylocladium AtrovircTis. (a) The fruiting organs of the fungus. (6) A portion of conidiophore with attached conidia. (c) The several-celled .spores, which usually put out a germ tul^e fron. the pointed end of the spore. (After Appel and Laubert.) colored areas. The relation of the sclerotia to the cells of the outer skin of the potato is shown in figure 4. \Yhen viable s})ores are placed in water at room tem])erature, they push out germ tubes in the manner shown in figure 1, c. Thes(^ may ])enetrate the periderm and later the ])arcnchyma, according to Frank. It has been reported by Johnson tliat sclerotia are even j^rochiced in the cells of the paren- chyma and that the fungus gi-ows readily on the cut surface of the potato. The wTiter has also observed the fungus fruiting on cut sur- ICii-. IL'TI SILVER SCURF, A DISEASE OF THE POTATO. 17 faces of ]-)Otatoes. As the fungus spreads in the tissues it loosens the surface lavers, causino; the wliite silvery color. Later the surface lay ers slouo-h off, carrying with, them numerous sclerotia that infect the soil. In moist conditions the fungus fruits in from four to five days after the spores have germinated. The conidiophores may spring either from the mycelium in the tissues or from a sclerotium, in the manner shown in figure 1, a. The conidiophores are dark brown, i'f Fig. 2.— Potatoes (Irish Cobbler variety) infected with Spondylocladium atrovirens. The lower right-hand tuber is irse from infection, while the other two show successive stages of infection. The diseased areas in this stage of development have not become shrunk and shriveled. rather tall, being about 120 microns long and 4 microns in diameter. (See fig. 1.) The spores are Dorne in rather irregular whorls on the upper portion of the conidiophore. The spores are dark browm, obo- vate, 5 to 7 celled, and about 40 microns in length by 8 in diameter. Nothing is knowai as to the longevity of the spores or the sclerotia. It is very probable that both are f|uite resistant and may therefore be important factors in facilitating the spread and continuance of the disease. It has been shown by Eichinger that the germ tubes and mycelium grow from the light, showing a negatively heliotropic reaction. 91517°— Cir. 127—13 3 18 ClBCULAR NO. 127, BUREAU OF PLANT INDUSTRY, The sclerotial stage was until rec(Mitly thought to be due to another fungus called Phellomyces sderotiopltorus. Black sclerotia, S to 16 microns in dianietcn', are formed more often in the surface layer of cells (fig. 4), although, according to Johnson, they may also be pro- duced in the dee))er layers of the periderm and parenchyma. From this obseryation it would seem that the fungus is ah\o to attack any part of the tuber, a fact tending to show its parasitic nature. Fig. 3.— Later stages of silvpr scurf infection in the same variety of potato as shown in figure 2. It is a very common occurrence to find such infections as shewn in the lower right-hand tuber. The ctTect of this disease on the potato is clearly shown in the upper tuber. The surfaces are shriveled, due to the loss of moisture and sloughing off of the protective covering. The fungus causing silyer scurf has been described as Spondylo- cladium atrovirens by Appel and Laubert. The spores are reported to be "4-8 (Meist 6-9) Querwanden, 7, 8-11, 9; 36-61, 5, ini Mittel IQ-AQ/j.." Clinton found that the species he studied had smaller spores. The WTiter has also found that the sj)orcs are not nearly as large as re])orted by Appel and Laul^ert. Wheth(n- this difl'ert^nce in the size of the spores is suflicient cause for considering the American species different from the European one is not clear at the present time. According to Clinton it has been suggested by Appel that there may be at least two forms, one haying larger s})ores than the other. A study of this matter is in progress. U'ir. 1-7 1 SILVER SCURF, A DISEASE OF THE POTATO. 19 This disease was called " Fleckenkranklieit " and also " Phellomyces- Faule" by Frank in 1897 and 1898, respectively. It can hardly be considered a spot disease, since it very often involves the whole surface of the diseased tuber. Recently it has been shown by Appel and Laubert that PlieUomyces sclerotiopJiorus, the fungus described by Frank, is only the sclerotial stage of S pondylocladium atrovirens. This fact naturally excludes the possibility of using the name applied to this disease by Frank. More recently the disease under consider- ation has been called " scab," " dry rot," '' dry scab," and '' scurf." It does not seem wise nor justifiable to apply the term "scab" to this disease, since we already have three or four different kinds of scab. Furthermore, Spondylocla- dium does not cause symptoms comparable to those produced by Oospora, Spongospora, and Chrys- ophlyctis. As is well known, all of these parasites cause hypertro- phy and cankers of the tissues in- fected. Such is not the case with Spondylocladium. On the con- trary, it causes the death and shrinkage of the attacked cells and tissues. The term "scurf" has been applied by Clinton and is doubtless a better name than any used previously, because it is less confusing and is more de- scriptive of the symptoms of the disease. The light, glistening, silvery discoloration of the sur- face of the washed potato due to the Spondylocladium infection is a very common and characteristic symptom. In view of this fact the name "silver scurf" is proposed for this disease. Fig. 4.— Sclerotia of silver scurf, (a) The surface view of aa infected area, showing the sclsrotia and their relation to the outer layer of cells. (6) An isolated sclerotium, much enlarged, in the early stages of germination. (After T. Johnson.) DISTRIBUTION OF THE DISEASE. Silver scurf has been known in Europe since 1871, when it was apparently first found and described by Hertz in Austria. In 1897-98 it was reported in Germany by Frank as causing* a sort of dry rot. Johnson in 1903 reported it as causing considerable dry rot in Ireland and the following year it was reported in England by Smith and Riea. Up to the present writing silver scurf has been reported from only one locality in the United States. Clinton found it in Connecticut "on a few tubers among the many varieties to be tested [Civ. IL'7] 20 CIKCULAK NO. 127, BUEEAU OF PLANT INDUSTRY. that year at the station." It should be noted that he beheves the disease was introduced into Connecticut on imported varieties. The disease was first found by tlie writer in the fall of 1912 in two barrels of potatoes shipped from western New York to Washington, D. G. No count was made, but it was estimated that at least 10 per cent of these potatoes was infected. Later it was found on Irish Cobblers bought in the local market early in December, 1912. These potatoes were raised in Maryland about 15 miles from Wash- mgton. Each tuber was examined and it was found that 56 per cent of this collection (2 bushels) was infected with silver- scurf. More recently the disease has been found on potatoes from Virginia, Ver- mont, Maine, Kansas, West Virginia, New York, Florida, and Wis- consin, which tends to show that it may be quite generally distrib- uted throughout the eastern half of the United States. This is espe- cially interesting as showing its rate of spread, since it was first observed in this country in 1907 by Clinton and could hardly have existed long prior to this date without being observed by patholo- gists. IS IT INJURIOUS? It has been quite definitely estabhshed by earlier workers that Spondylocladium does not attack the fohage of the potato plant, but is confined cliiefly to the potato tuber. Although the morphology of this fungus has been beautifully pictured and well described, very little is known regarding its eft'ect on the tuber further than a few general statements occurring in the literature, none of which are based on experimental evidence. It is not probable that silver scurf is as destructive as our common Oospora scab, yet the following statements tend to show that some damage does occur from this disease. According to Frank, silver scurf has the same effect on the potato as the Fusaria. This comparison is hardly justifiable in the fight of our recent contributions on the various parasitic species of Fusaria, although it conveys some idea as to its destructiveness. Appel and Laubert are of the opinion that Spondylocladium has Uttle destruc- tive effect on the potato. It has already been pointed out that the fungus studied by Appel and Laubert may be different from the one common on the potato in America. Johnson, of Ireland, befieves that "it is a true parasite, and apparently may do considerable unsuspected damage to the potato crop in Ireland. It may give rise to a skin disease or 'scab' of the potato tuber, and in a more advanced stage of attack to a 'potato rot'." Massee recommends that infected tubers should not be used for seed purposes. By Bohutinsky it has been stated that a species of llelminthosporium is the cause of leaf-roll. The fungus reported as causing leaf-roU has since been identified by [Cir. 127] SILVER SCURF, A DISEASE OF THE POTATO. 21 Appel as Sporuhjlocladium atrovirens. This is also quite apparent from Bohutinsky's figures. It does not seem very probable that silver scurf is the cause of leaf-roll, but it may well be that it causes abnormal plants in certain instances. The fact that it disfigures the tuber and causes abnormal shrinkage is alone sufficient to involve considerable loss where tubers are to be used for human consump- tion after a considerable period of storage. It is not without destruc- tive effect on the early varieties as well, and especially with red- skinned ones. A shipment has come to the writer's attention in which marked disfiguration occurred in the Bhss Triumph variety grown in the South for the early market. The tubers in some cases were so badly infected that the red color, so characteristic of the sound tuber, was no longer in evidence. In such cases considerable loss occurs, due to the sorting and to the decrease in market value of the infected tubers. The big question in connection w^th silver scurf has to do with the effect it wiU have on the tubers used for seed purposes. What will be the effect on the >aeld and general vigor of the crop grown from seed infected mth this disease ? What effect will it have on germi- nation, and will the wounds or lesions due to Spondylocladium increase the susceptibility of the parent tuber to the ravages of wound para- sites? Likewise, the relation of the fungus to the soil and the method of spread of the disease in storage are open questions. These are problems that can not be answered at present, but wliich merit careful consideration, especially where the disease is very prevalent and the tubers are held in poor storage for a considerable length of time. POSSIBLE METHODS OF CONTROL. It is of special interest to know whether tliis disease, confined as it is chiefly to the periderm, can be controlled by the well-known formalin treatment used for the Oospora scab. It is claimed by Johnson that such is the case. He treated tubers for one hour in an 0.8 per cent formahn solution (about 2 pints to 30 gallons of water) and planted the same. Other diseased tubers not treated were planted as checks. When the pota^toes were harvested in the fall it was found that the treated tubers were free from disease, while those from the checks were infected with the brown-mold stage of Spondylocladium, indi- cating that the 0.8 per cent formahn solution either killed or materially inhibited the growth of the fungus. The effect of the formahn and corrosive-sublimate treatment as a means of destroying silver scurf has also been tested in the laboratory of Cotton and Truck Disease and Sugar-Plant Investigations. Two varieties of potatoes, the Irish Cobbler and the Green Mountain, were washed and soaked in formahn or corrosive sublimate. Follow- [Cir. 127] 22 CIRCULAR NO. 127, BUREAU OF PLANT INDUSTRY. ing this treatment the specimens were placed in moist chambers under conditions favorable for the fungus to fruit. Iji every case viable spores were produced from 2 to 12 days after treatment. The germinating capacity of the spores was determined by placing them in water for 24 hours. In every case the spores germinated. As shown in Table I, neither corrosive subhmate nor formalin used twice as strong as in common practice for a period of five hours killed the fungus. Table I. — Effect of formalin and corrosive sublimate on silver scinf. Variety tested. Number treated. Fungicide used. Result. Date. Name. Strength. Time of treat- ment. Period in moist ehamljer. N'iable spores. Jan. 30 f 50 Irish Cobbler \ [ 25 ,, ^ ■ f 10 Corrosive s u 1j 1 i - mate. Formalin do Percent. 0.2 .8 .8 Hours. 2 2 2 li 5 5 4 4 3 1§ 19 5 Days. 2 2 12 12 4 4 2 f> 10 10 Present. Do. Do. Feb. 10 Green Mountain... ,do- i 10 f 10 i 10 f 10 i 10 f 10 \ 10 20 20 Corrosive s u b 1 i - mate. Formalin Do. Do. Feb 11 Corrosive s u b 1 i - mate. Formalin Do. Do. Irish Cobbler do Feb. 12 Corrosive s u 1) 1 i - mate. Formalin Do. Do. Feb 14 Corrosive s u b 1 i - mate. Formalin Do. Do. Mar 20 F.nrpka Mar 21 .do do Do. These data suggest that the ordinary formalin and corrosive- sublimate methods in vogue for controllmg Oospora scab do not kill Spondylocladium. This is not surprising in view of the fact that the fungus produces numerous sclerotia and penetrates all parts of the periderm and in severe cases even extends into the parenchyma. Johnson used an 0.8 per cent solution of formalin and succeeded m preventing the reappearance of silver scurf. An 0.8 per cent solu- tion of formalin is twice as strong as that commonly used in this country for Oospora scab. In the test on January 30, 1913, recorded in Table I, a double strength of corrosive sublimate (about 4 ounces to 15 gallons of water) killed the tubers, but the fungus developed viable spores after 7 days. On March 20, infected tubers were soaked 19 hours in an 0.8 per cent formalin solution and immediately placed in a moist chamber. The fungus produced viable spores after 10 days. It should also be noted that many of the eyes of the tubers w^ere killed by this treatment. Neitlier does dr}' heat at 50° C. for 5 hours kill the fungus. This was learned in connection \\dth numerous experiments planned to determine the effect of dry [Clr. 127] SILVEE SCURF, A DISEASE OF THE POTATO. 23 heat on tubers infected with PJiytophthora infestans. Spondylocla- dium fruited from 4 to 10 days after the tubers had been subjected to 50° C. for 2 to 10 hours. If the potatoes were kept at 50° C. for longer periods their germmating capacity was injured. This tends to show that the viability of the fungus is even greater than that of its host. Experiments are in progress to determine further the value of formalin as a means of controlling this disease. RELATION TO THE SEED-POTATO INDUSTRY. In order to gain some idea as to the amount of sUver scurf present on potatoes being used for seed purposes tliis spring, several days were spent examhihig the seed stock being planted m the vicmity of Norfolk, Va. A shipment of over 500 barrels raised in northern Vermont came to the writer's attention. These potatoes were shipped to Norfolk last November and stored at a temperature rangmg from 40° to 65° F., accordmg to the statement of the owner. Some Phytophthora rot was in evidence in tliis stock. The combmation of high temperature and moist condition, the latter due to the soft- rot, favored the development of Spondylocladium. Both the brown- mold and the sclerotial stages were readily detected, especially the former. Three barrels were sorted with the following results : First barrel (by measui-e), infection with silver scurf 40 per cent. Second barrel (by measure), infection with silver scurf 2 per cent. Third barrel (by measure), infection with silver scurf 50 per cent. It was impossible to examine every potato in any considerable number of barrels, so 17 other barrels from the collection of over 500 were dumped separately and the amount of silver scurf estimated. This was not especially difficult, because of the prevalence of the brown-mold stage, wliich is quite conspicuous when well developed. The amount of silver scurf ranged from about 1 to 90 per cent. When the amount of soft-rot was great the percentage of silver scurf was less, due, it is believed, to the excessive moisture present, a con- dition unfavorable for the rapid spread of the fungus. In 10 of the barrels exammed the amount of silver scurf ranged from 25 to 90 per cent and in the remauiing 10 it varied from about 1 to 25 per cent. Another collection, consistmg of 25 ban-els of potatoes which had just arrived from Ai'oostook County, Me., showed about 25 per cent of silver scurf. It was strikingly noticeable in tliis collection that the disease was not as far advanced as in the earlier collection ex- amined, and only the sclerotial stage was in evidence. Still another collection from Maine needs mention. It consisted of 15 barrels grown in the south-central part of the State. This collection was remarkably free from late-blight, but showed 35 per cent of silver [Civ. 127] 24 CIRCULAR NO. 127, BUREAT OF PLANT INDUSTRY. scurf. Ill three of the barrels that were sorted the following amount of Spondylocladmm were found: First barrel (by measure), infection with silver scurf 30 per cent. Second barrel (by measure), infection with silver scurf. ... 40 per cent. Third barrel (by measure), infection with silver scurf 35 per cent. Tliree of the other barrels were emptied and the amount esti- mated. It was approximately the same as in the tliree barrels sorted. The infection in this collection had progressed farther than in the shipment examined from Aroostook County. Tliis was pos- sibly due to the different storage conditions to which the two lots of potatoes had been subjected. Small collections of other lots of seed potatoes from IMaine on which silver scurf was also present were brought to the writer's attention, but the c^uantity examined was insufhcient to make any just estimate as to the amount of this disease. Although only comparatively few lots of seed have been examined, yet it is sufficient to show that silver scurf is a common disease on potatoes in Maine and Vermont, two States that furnish a consider- able amount of seed to various sections of the United States, and smce it is present on the seed going out from these two States it is possible that silver scurf is quite generally distributed in the eastern half of the United States. BIBLIOGRAPHY. 1871. Harz, r. O. Einige neue Hyphomyceten Berlin's und Wien's nebst Beitragen zur Systematik derselben. Bulletin, Societe Imperiale des Naturalistes, Moscow, t. 44, pt. 1, p. 129. 1897. Frank, A. B. Kampfbuch gegen die Schadlinge unserer Feldfriichte. Ber- lin, p. 182-185, 197-198, fig. 33, 37. 1898. Frank, B. Untersuchungen iiber die verschiedenen Erreger der Kartoffel- faule.. Berichte, Deutsche Botanische Gesellschaft, Bd. 16, Heft 8, p. 280-281. 1903. Johnson, T. Phellomyces sclerotiophorus, Frank: A cause of potato seal) and dry rot. Economic Proceedings, Royal Dublin Society, v. 1, pt. 4, p. 161- 166, pi. 2-3. 1905. Appel, Otto, and Lai^bert, R. Die Konidienform des Kartoffelpilzes Phel- lomyces sclerotiphorus Frank. Berichte, Deutsche Botanische Gesellschaft, Bd. 23, Heft 5, p. 218-220. 1907. Appel, Otto, and Laubert, R. Die Konidienform und die pathologische Bedeutung des Kartoffelpilzes Phellomyces sclerotiojihonis Frank. Arbei- ten, Kaiserliche Biologische Anstalt fiir Land- und i'Vjrstwirtschaft , l^d. 5, Hefty, p. 435-441, pl.ll. 1908. Clinton, G. P. Scurf, Spondylocladium atrovirens Harz. Connecticut Agri- cultural Experiment Station, Annual Report, 31/32, [1907]-1908. p. 357-359. 1909. Eichinoer, A. Zur Kenntnis einiger Schalenpilze der Kartoffel. .\nnales Mycologici, v. 7, no. 4, p. 356-364, 3 fig. 1909. Massee, George. Dry scab of potatoes. (Spondylocladium atrovirens, Harz.) Royal Botanic Gardens, Kew, Bidletin 1. p. 16-18, 3 fig. 1910. BoHUTiNSKY, G. Beitrage zur Erforschung der Bhittrollkraiikhcil. Zeit- schrift fiir das Landwirtschaftliche Versuchswe.se n in Oesterreicli, Jalirg. 13, Heft 7, p. 607-<)33. 3 fig. 1910. Ma.s.see, George. Diseases of Cultivated Plants and Trees. London, p. 478-480, fig. 142. 1912. Eriks-son, .Jakob. Fungoid Diseases of Agricnltiinil Plants. Tr. from the Swedish l)y Anna Molander. London, p. tl2. [Clr. 127] [Cir. 127— C] THE DASHEEN, A ROOT CROP FOR THE SOUTHERN STATES/ By Robert A. Young, Scientijic Assistant, Office of Foreign Seed and Plant Intro- duction. INTRODUCTION. While Mr. O. W. Barrett was connected with the Porto Rico Agri- cultural Expeiiment vStation previous to 1905, he brought together and grew a large collection of the tuberous-rooted aroids which are important food plants of the Tropics. He drew attention to the pos- sibilities of these aroids in a bulletin published in 1905,^ and when he entered the Office of Foreign Seed and Plant Introduction he assem- bled a still larger collection from different parts of the world. Several field trials of them were made at Gotha, Fla., and Gough, S. C. The results of these early experiments in the United States were published in 1910.^ As a result of these preliminary trials, in which, naturally, many of the introduced varieties failed, one of. the forms, the dasheen, proved to be of unusual promise. This paper deals with some of the more important features of the field experiments with the dasheen under the writer's direct super- vision, as well as some greenhouse experiments in the production of blanched shoots, and with the results of many experiments in the cooking and preparation of the shoots and tubers for the table. In 1909, from a small experimental plat grown near Charleston, S. C, it became apparent that the dasheens, which are closely allied to the taros of Hawaii, China, and Polynesia, were well adapted for culture in certain of the moist lands of the South. Since that time the endeavor has been to propagate a stock of the best varieties sufficient to make possible their distribution on a large scale. Certain varieties secured originally from the island of Trinidad and other parts of the West Indies were found to be satisfactory in 3deld and to be of higher quality than others in the collection. These are the varieties to which particular attention is now being given, though many others are under investigation. 1 Issued May 17, 1913. 2 Barrett, O. W. The yautias, or taniers, of Forto Rico. Porlo Rico .Vgricultural Experiment Station, Bulletin 6, 27 p., 4 pi., 1905. 3 Barrett, O. W., andCooli.O. F. Promisingroot crops for the South. U. S. Department of Agriculture, Bureau of Plant Industry, Bulletin 164, 43 p., 1910. [Cir. 127] 25 26 CIECULAR NO. 127, BUREAU OF PLANT INDUSTRY. The origin of the word ''dasheen" is somewhat obscure, but Mr. Barrett, who has spent some time in Trinidad, as well as in other parts of the West Indies, states that it originated in Trinidad and suggests an explanation which seems at least probable. Two pos- sible derivations are given: "de la Chine" or "da Chine" (the latter pronounced dah-sheen, being the form in the patois of the French West Indies), meaning "from China"; also "des Indes" (pronounced daze-eend, or daze-and), meaning "from the Indies." Mr. Barrett believes the former to offer the more likely explanation. Fig. 1.— a field of dasheens'as it appears in September. These broad-leaved plants when full size stand 4 to 6 feet high and shade the ground completely. The seed tubers are planted in March; the harvest takes place in October. This planting is on hammock land at the United States Plant Introduction Field Station, near Brooksville, Fla. The dasheen is also known in various parts of tropical America under the names "malanga," "eddo," "coco," "taya," and "tanier" (also spelled "tannia" and "tanyah"). These names are likewise often applied to the yautia and taro. DESCRIPTION. Dasheen plants are members of the botanical family Aracea^, to which belong also the calla, the Indian turnip, and the caladium, or elephant's-ear. The dasheen belongs to the genus Colocasia, but there has not as yet been sufficient botanical work done on this group [Cir. 127] THE DASHEEN, A ROOT CROP FOR THE SOUTHERN STATES. 27 to make it possible to determine the species. The leaves of the dasheen are shield shaped (figs. 1 and 2), like the caladium and the tanier of the South, and contain the same acrid principle that characterizes the Indian turnip and most other plants of this family. They should never be tasted raw. The tubers of the most promising Trinidad Fig. 2.— a hill of the Trinidad dasheen. One dasheen plant will cover a square yard of ground and pro- duce 4 to 10 pounds of tubers on good rich soil. This variety is one of the best flavored yet introduced. varieties are free from this acridity even in the raw state, but because of the possibility of tubers of an acrid variety bemg mixed with these it is best never to taste them uncooked. In cases of the accidental tasting of acrid tubers or leaves, lemon juice in a little water is found to alleviate the ill effects. [Cif. 127] 28 CIRCULAR NO. 127, BUREAU OF PLANT INDUSTRY. If dasheeiis are handled in water in scraping or paring them for cooking, a level teaspoonf ul of sal soda should be added to each quart of water. The outer part of the tubers contains an irritant that causes the hands to sting in somewhat the same way as the mouth and throat from the eating of raw, acrid leaves or tubers. The hands are affected m tliis way even in the case of tubers that are not acrid to the taste. If water is not used while scraping them, it is best to wash the hands afterwards in soda water of the strength mentioned. The dasheen corms and tubers ^ are similar to the potato in com- position, but they contain less w