BULLETIN 232 MAY 1930 EXPERIMENTS WITH LEGUMES IN ALABAMA By R. Y. BAILEY J. T. WILLIAMSON J. F. DUGGAR AGRICULTURAL EXPERIMENT STATION OF THE ALABAMA POLYTECHNIC INSTITUTE M. J. FUNCHESS, Director AUBURN, ALABAMA AGRICULTURAL EXPERIMENT STATION *STATION STAFF Bradford Knapp, B. S., LL. B., D. Agr., President, 1928 M. J. Funchess, M. S., Director of Experiment Station, 1909, 1924 W. H. Weidenbach, B. S., Secretary, 1925 P. O. Davis, B. S., Agricultural Editor, 1916, 1921 Mary E. Martin, Librarian, 1912, 1918 AGRONOMY M. J. Funchess, M. S., Head Professor of Agronomy, 1909, 1921 J. W. Tidmore, Ph. D., Professor of Soil Chemistry, 1928 L. D. Baver, Ph. D., Associate Professor of Soil Chemistry, 1929 H. B. Tisdale, M. S., Associate Professor of Plant Breeding, 1913, 1917 J. T. Williamson, B. S., Associate Professor of Agronomy, 1911 R. Y. Bailey, B. S., Assistant Professor of Agronomy, 1927 D. G. Sturkie, M. S., Assistant Professor of Agronomy, 1925 G. D. Scarseth, M. S., Assistant Professor of Soil Chemistry, 1928 **W. W. Pate, M. S., Assistant Professor of Soil Chemistry, 1929 G. H. Jester, B. S., Assistant in Agronomy, 1928 W. D. Lucas, B. S., Assistant in Agronomy, 1928 E. L. Mayton, B. S., Assistant in Agronomy, 1929 J. A. Naftel, B. S., Assistant in Agronomy, 1928 J. W. Richardson, B. S., Assistant in Agronomy, 1929 J. R. Taylor, B. S., Assistant in Agronomy, 1929 C. J. Rehling, B. S., Assistant in Agronomy, 1929 H. W. Bennett, B. S., Graduate Assistant, 1929 ANIMAL HUSBANDRY, DAIRYING AND POULTRY J. C. Grimes, M. S., Head Professor of Animal Husbandry, Dairying and Poultry, W. D. Salmon, A. M., Research Professor of Animal Nutrition, 1922, 1927 G. A. Trollope, B. S., Professor of Poultry Husbandry, 1924, 1929 N. B. Guerrant, Ph. D., Research Associate Professor of Animal Nutrition, 1927 Dale F. King, M. S., Assistant Professor of Poultry Husbandry, 1930 C. T. Bailey, B. S., Superintendent, Poultry Farm, 1929 BOTANY AND PLANT PATHOLOGY W. A. Gardner, Ph. D., Head Professor of Botany and Plant Pathology, 1917 J. L. Seal, Ph. D., Associate Professor of Plant Pathology, 1929 G. L. Fick, M. S., Assistant Professor of Botany, 1926, 1928 E. V. Smith, B. S., Graduate Assistant, 1929 ECONOMICS (AGRICULTURAL) J. D. Pope, M. S., Head Professor of Agricultural Economics, 1924, 1926 B. F. Alvord, M. S., Associate Professor of Agricultural Economics, 1929 C. G. Garman, IB. S., Assistant Professor of Agricultural Economics, 1928 ***H. T. Wingate, B. S., Assistant in Agricultural Economics, 1926 E. E. McLean, M. A., Assistant in Agricultural Economics, 1929 Carl F. Clark, M. S., Assistant in Agricultural Economics, 1929 Edith May Slights, Statistical Assistant, 1926 ENGINEERING (AGRICULTURAL) M. L. Nichols, M. S., Head Professor of Agricultural Engineering, 1919 J. W. Randolph, M. S., Assistant Professor of Agricultural Engineering, 1920 Arvey Carnes, B. S., Assistant Professor of Agricultural Engineering, 1923, 1929 Ellis Diseker, B. S., Assistant in Agricultural Engineering, 1928 T. N. Jones, B. S., Graduate Assistant, 1929 ENTOMOLOGY J. M. Robinson, M. A., Head Professor of Zoology-Entomology, 1919, 1929 H. S. Swingle, M. S., Associate Professor of Entomology, 1929 L. L. English, Ph. D., Associate Professor of Entomology, 1928 F. S. Arant, B. S., Assistant in Entomology, 1926 FARM MANAGEMENT J. F. Duggar, M. S., Professor of Farm Management. 1897, 1922 HOME ECONOMICS RESEARCH Louise P. Clanton, M. A., Head Professor of Home Economics, 1927 A. L. Sommer, Ph. D., Associate Professor of Home Economics, 1930 Helen Dumond Herren, A. M., Assistant in Home Economics, 1928 HORTICULTURE AND FORESTRY C. L. Isbell, Ph. D., Head Professor of Horticulture and Forestry, 1917, 1929 W. I). Kimbrough, Ph. D., Associate Professor of IHorticulture, 1925, 1929 O. C. Medlock, M. S., Assistant Professor of Horticulture, 1928 R. W. Taylor, M. S., Assistant Professor of Horticulture, 1924 P. L. Wright, B. S., Graduate Assistant, 1929 SUBSTATIONS Fred Stewart, B. S., Superintendent, Tennessee Valley Substation, 1929 R. C. Christopher, B. S., Superintendent, Sand Mountain Substation, 1929 J. P. Wilson, B. S., Superintendent, Wiregrass Substation, 1929 K. G. Baker, B. S., Superintendent, Black Belt Substation, 1930 Otto Brown, M. S., Superintendent, Gulf Coast Substation, 1930 * 1920 Station Staff as of March 1, 1930 ** Assigned by the State Department of Agriculture and Industries ***In cooperation with U. S. Department of Agricultural Economics and Georgia Experiment Station EXPERIMENTS WITH LEGUMES IN ALABAMA By R. Y. BAILEY Assistant Agronomist J. T. WILLIAMSON Associate.Agrononist Research Professor inFarm Management.. J. F. DUGGAR BULLETIN 232 -n I MAY MY13 1930 Table of Contents I. II. INTRODUCTION ................... 3 . 4 INFLUENCE OF LEGUMES ON CROP YIELDS 4 A.-The Value of Legumes in Rotation ................ B.-The Value of Legumes in Continuous Crop..... . ... ...... .. 14 ping System s ......... RESIDUAL EFFECT OF LEGUMES ............................ 18 A.-The Effect of Vetch and Annual Yellow Melilotus Turned in 1925, 1926, and 1927 on the Yields of Corn and Cotton in 1928 ....... 18 B.-Effect of Kudzu on the Yields of Succeeding ...... .. 21 Crops .................. CULTURAL EXPERIMENTS WITH LEGUMES ...... A.-Effect of Fertilizing Materials on the Growth of Legumes ...................................... B.-Time, Rate, and Methods of Seeding Winter L egum es ................. ......-............ ...... ........... C.-Growth and Yields of Winter Legumes ........ D.-Relative Earliness of Winter Legumes .............. III. IV. -21 21 27 30 35 V.. VI. NITROGEN CONTENT OF WINTER LEGUMES ............ 36 A.-Per Cent of Nitrogen in Green Matter .............. 36 THE PRODUCTION AND HARVESTING OF WINTER LEGUM E SEED ...................... ............ .... A.- Regular Variety Test ---.............. ............... B.- General Field Planting ................ ................ C.- Spring Planting ...................... .................... COLD, INSECT, AND DISEASE INJURY TO WINTER LEG U M ES ............. - . -.. ....... .... ..... ...... A .- W inter Injury ......................... ........ .... ............. B.-Variable Response of the Same Species to Low Temperatures ...................-----------------------.......................... C.-Relative Susceptibility to Insect Injury and Disease .......................-------------------------------- . ........................----------------------------------36 36 37 37 37 37 41 42 43 VII. VIII. SUMMARY Experiments With Legumes in Alabama ment Station was that of soil fertility. This phase of work has received the attention of this station for more than thirty-five years and is still considered a problem of major importance. Many experiments have been conducted for the purpose of studying the effects of legumes turned under on the yields of succeeding crops. Other experiments have been conducted for the purpose of determining which legumes are best suited to Alabama conditions and for the purpose of studying some of the problems involved in the successful growing of these legumes. Prior to 1910 this station conducted a number of tests with winter legumes in co-operation with Alabama farmers. For some reason legumes planted in these tests failed more often than the legumes in the tests at Auburn. When the Local Experiment Fund was made available in 1911, co-operative tests which were supervised by a station representative were started. The purpose of these earlier tests was to determine why legumes were grown less successfully over the State than at Auburn and to find out which legumes were best adapted to Alabama conditions. The results of this earlier work emphasized the need of phos- O INTRODUCTION NE of the first problems studied at the Alabama Experi- phorus and the inoculation of legumes by the proper bacteria. Consequently, all of the experiments with winter legumes conducted by the Alabama station have received phosphorus; and if the land had not already been properly inoculated bacteria were applied. This bulletin records results of experiments with legumes by the Alabama Experiment Station from 1896 through 1929. With the exception of analyses of legumes, the data reported have been obtained through investigations carried on in the field. The greater part of the results herein reported has been obtained from experiments at Auburn; co-operative experiments, however, have afforded a part of the data as well as giving valuable leads for planning experiments on the station farm. Experiments were conducted to study the influence of legumes on the yields of succeeding crops; residual effect of legumes; influence of fertilizers and lime on yields of legumes; time, method, and rate of seeding legumes; yields of legumes at different stages of growth; nitrogen content of legumes; seed production of legumes; and the effect of low temperatures on legumes. INFLUENCE OF LEGUMES ON CROP YIELDS the((l~l croppg The Value of Legumes in Rotation.-The xvaluie of le-nies in systemii h as been sh own by a long, coni111 ne experi- menit condutctedl at the Alab~ama Experiment Station; the average resuilts liy teln year peinol are g'iveln in Table 1. Plots 1 aiid 2 in this expelimenit wrere plantedl to corni each year since the exp~erimnt was starlted. C owpeas were planited ini the corn midd~tles oil Plot 1, at the last cultiv ation from the liegililig of the ex perimnit itli 1 925, a ft er which fall pl)ant ed xetc h was suibst it ntedl foi cowlea pNl~ anted inr the suimm er. All cow peas and retch were plowed olider for the next crop of corn. No leglnies wele planited on Plot 2. The ax erage results on Plot 2 show a steady detclinie iii yieldl. Plot 1, wxhich has been pilanitedl to (cor1 eachi year' withi legunies in the corni mlidd~les, has produced slightly more corn diiriiig the last ten years than (hirinig the first te11 reals of the expeirimenit. 'Ibis pilot is niowr irotitcing molet Ihan twice as much corn ias Plot 2. A study of the y'ields foir the finst tweiitr reals shows that cowpeasit planited in the coin mitddles didI not maintain tht rield oft corn on P~lot 1. '[he coxx'ptas. that xrere tuinetd under on this plot resulted in SeeT.he I Ivgtum'crop uII earn1 %Ohl .1011 \\thotit ttti\( u 2It :l IB'(lRE 2.-on left, Iithlit b VldedC~ annually. Corn on right folloWitig 1e911Cgnn-i (Plots I and 2, x it hout leguntcs, corn produced 8.9 bushels per acre. See Table 1. Old Rotation Experiment. Photographed June 20, 1929.) b ushels of (0111 per acrie over Plot 2 dIuing the second~ ten-year p)eriod, 1906-1915. Vetch was sublst ituted for cowlpeas in 1925 ;sinc e that time the average yieldl on lPlot 1 has incrieased slightly above that for the fir'st ten years. The valtie of legumes was further shown by the results ob)t ained from otth er plots in this experiment. Plot 6 has grown cottonl continuouitisly withotit legumes. The yield of seed cotton diec linied from 803 pounds ini the first ten-y ear pteriodl to 8349 pouni lds per acre as the av erage for the ten-year period trom 192(0 to 1929. inclusive. Plots :8 andl 8 w'ere plantedl to cttoni each vear: vetch w-as lplanted in the fall to be turned under the followxi ng sprintg for cotton. Ijuring the first ten-year period these tc twot plots produ ed an average yield of 813 pounds of seed1 cotton1 per acre; an average o1' 678 pounds the second ten years; anid an axverage ot 756 ptounds in the lasi ten years, 1920) to 1929. inclusixve. ( )tn Plots 5 aittl 9 at two-year rotationt of cttltton, with vetch planted ini the cotton middles in the fall, and coxwpeas1 was pc followed from 1896 until 1923. Vetch and cowpleas wer Ilowedi ntder for suicceedling crops. In 1923, and thereafter, the cowp~eas xvere cut tor hay and followed by xetch in the fall to be turined the following spring for cotton. This treatment resulted ant inWlease (1f 6 TABLE 1.-The Effect of Legumes Turned Under on the Yield of Succeeding Crops. Fertilizers 5. Lbs. per acre Plot No. 1 2 6 3 & 8 5 & 9 Cropping system Superphosphate Kainit 160 160 160 160 160 Peas at last Corn each year. cultivation 1 Corn each year, no peas Cotton continuously, no legumes Cotton and vetch continuously Vetch as cover crop Cotton and vetch. Cowpeas 2 6nSeed 3 yr. av.6 yield of vetch. Lbs. green matter per acre Average yields of crops Last 10 First 10 Second 10 years years years 1896-1905 1906-1915 1920-19294 Corn--bushels per acre 19.2r 17.13 803 16.2 19.4 10.2 8.2 Cotton-pounds per acre 573 349 678 9583 756 1,041 160 160 160 160 160 12,918 9,570 10,904 813 890 1 Changed to corn and vetch in 1925. 2 In 1923 and after, peas cut for hay and followed by vetch. 3 Only 9 crops. 4 Cotton and corn failed on all plots in 1925, due to drought. 5 In the fall of 1921, 400 pounds superphosphate were applied same half of each plot in the fall of 1922. The east half of all plots 1923 to equalize the phosphate application. In the fall of 1923, and have been applied to the whole of each plot. 6 Yields for 1926, 1927, and 1929. 6 Ten year average of 9 crops. to the west half of each plot. This was repeated on the received 800 pounds of superphosphate in the spring of each fall since, 400 pounds of superphosphate per acre in an increase of 151 p~ounds of seedl cotton per acre for the last en years over the average for the first ten years. Each plot in the experiment discussed above was fer'til izedl a like so that the dlifterences obtainedl in vieldIs were du to the le Ilegumes5 that were turned u ndler. Vetc h crops were p~oor prior to 192:1. Beginning ini 1 923, and thereafter, eac h p lot h as receiv ed an application of 400 pounds of superphosphate per' acre in the fall. Since th is a(dditijonal p~hosphorus treatmnient was started1, the x etch crops h ave b~een ent irelyv sat is factory. A comn p~arison of the cotton vie lds on Plfots a nfl 8 for the second( tenl x'ears wxith the vieldls for the last ten years shows that the increasedI amount of \ etch turned under materially increased the yieldIs of cotton. This is espcially interesting since the ax erage for the last ten years is a ten year ax erage of' only nine crop~s. The xvaloe of legumes in the croppfinig sy stem and the effect of phosphorus on legumes are clearly showxn by the (lata fpresentedl in Table 2. In this experiment cowpeas and xvetch were used1 as much as possible in a three-yeair rotation of corn.ot ! Y :4 ,t , p . o " ' ryw F'IGULRE Hiny vetch whinch pnroduced 1:;,. )4 pounds of green 1inateriai per acre following corn in a three-year rotation on Norfolk sandy lom soil. ( Plot A, Cullars~ Rotation. Photographed andl sampled April 18, 1925.) See Table 2. TABLE 2.-Influence of Legumes Turned Under on the Yields of Cotton and Corn Grown in a Three Year Rotation. Average yields last 9 years Plot ili eatment cropping system* Summer and winter Seed cotton. Pounds per acre Increase due to Seed cotton. Pounds per acre legumes Corn. Bushels per acre Corn. Bushels per acre A B C 240-Superphosphate 50-Muriate of potash 240-Superphosphate 50-Muriate of potash None 514 32.0 397 24.2 None 117 7.8 11.7 20.9 123 8.2 Plot A, 9,661 and Plot 1, 2,311 pounds per acre. None 98 Summer and 1 None winter 221 * Two year average yields of green matter turned under were: 8 4: /- SIiI ,' 1. ('orn in a three-yeal r otation. Icgtiies produred bu'snhels per acre gtliii , iuotieed 1.5 buh els( lier acre. ( Phtogap~hed June 21), l12. ) See 'Table The plot on the right, follwing the lot oil the left, wxithout le'Nots A and 1-i,('tihla Rotation. 2. toni.dlan oat s. Plots A a nd I; were ferti lizedl exactly alike, the oiily ifferencie in treatment being that Plot A had summer avid wiit ci legumfles in the rotation. P'lots A andil 1 had leg.umes in the rotation, but Plot 1 receivedl no phosphate andl potash. ]Plot A. w'hich had legumlns in the rotation. productiedl mlore than four times as much cotton and corn during the last nine yearis as Plot I; whirch had no legum es. A cornparison of the yields on Plots A. P, C, and I sho-ws that phosphate and potash allonie did not increase the yield of corn and illeleasedl the ield of cotton onily 19 pounds of seedl cotton per acre. When applied in the croppiig system with legumes, phosphate and p~otash resulted in an increase of 11 bushels of corn _ ndl 29:_ p'ounds ot seedi cotton per acric. In a nothe eixplerimlent conducted at the Alabama Experiment Staition, stale manure, nitrlate of soda, and winter legumes as. soy ices of niitroI gen for cotton andi corn were compared. This ex peiment was started in the fall of 192-1. Each treated p lot ini t he e xperi ment recei ved essentialIly the same phosphorus an p11lotash . The stale manure plots receive f'1ire tonls of manure per acre each spinvg. The manure wvas applied broadcast andI tu'ned( iind(1e1. The nitrate of sodla pilots rec eived 325 p~ounds of ntitrate of sodla per acre annunal ly. The vetch plo0ts received one toln of ground limestone per acre every fiv'e years for the benefit of the v etch, which was allowedl to grow until about A pril 1 before becing turnedi under. The experiment wvas dividedl into two sections on which cotton andl corn were alternated. " ~ ~> ' kV W. L. ,a,9: S49 / Ae ~f:. . r yr / yisz ;pz 4, i'H;ici' E 5. U (Plot a il, -1, ttrii I .2' II tt t I i r }l t t I' h anele. Allannie- N unte-A\etel E~pevlnwnt. tI(; I' G. Co('tton, Foloing the mnanatha \elth Shtwaa in lignac 5, whicho a iptttutaed 1(i pIotina of seal totton pet-ane . (Plot .1, Maanat-NitrateVetch IEftttiattat. P'htogaplhtad Septatete s, 19127.) See Tall :;. VIGtURE Th'is rp proill a t \ C:\ o otation. 7. (4100 \xithoiit legunim' Tis is a compainion plot to (luced 426 pounds o1 seed cotton per acr. (Plot 5, :tlanun oN itrate-Fetcb Experiment. the one s~hown in Figure G. Photographedl Septembe~r K, 11P27.)I See Table 3~. The average yields for five years and other dietails of this experiment are presented in Table 3. IPIots: 4 and 9, on which vetch was turned under, have produced slightly more cotton and 4.8 bushels of corn per acre less than plots which receiv ed 3~25 pounds of nitr'ate of sodla per acre. Stable manure produced 240o poulnds more seed cotton andI 7.4 bushels more corn per acre than v etch. Vetch was killed by cold in 1928 andI destroyed by rabbits on Plot 9 in 1929. The avecrage yields rep~resenlt the results of four crops of v etch on Plot 4 and three crops on Plot 9 as compared with fire apl~llications of manure and nitrate of soda on Plots 2. 3, 7, and 8. TABLE 3.--Influence of Stable Manure, Nitrate of Soda, and Vetch on the Yields of Cotton and Corn. Fertilizer treatment. Pounds per acre Five-year average*** yield of vetch. Pounds green matter per acre Yields per acre Five-year average 1925-1929 Pounds 1 & 5 2 3 No fertilizer Tons manure Superphosphate Nitrate of soda Superphosphate Muriate of potash Vetch* 600 Superphosphate** 100 Muriate of potash No fertilizer 5 325 200 100 Tons manure Nitrate of soda Superphosphate Muriate of potash 5 400 325 600 100 331 1,471 1,140 Average increase over checks seed cotton Plot 1,222 891 4 6 & 10 7 8 7,007 1,231 Bushels of corn per acre 7.9 38.3 900 30.4 35.7 27.8 Vetch* 9 400 Superphosphate** 100 Muriate of potash 6,855 30.9 * One ton of lime per acre was applied to Plots 4 and 9 every 5 years. ** Plots 4 and 9 received 400 pounds of superphosphate and 100 pounds of muriate of potash per vetch was planted. The plot on which cotton was planted received 200 pounds of superphosphate before the spring. *** Monantha vetch on Plots 4 and 9 was killed by cold in 1928. Austrian winter peas that were in January produced only 1900 pounds of green matter per acre. Rabbits darmaged vetch and Austrian so severely in 1929 that only 809 pounds of green matter per acre were produced. 12 23.0 acre in the fall when cotton was planted in planted on these plots winter peas on Plot 9 hIxperinitnt. P'hotographed .lu1 11 19G)Sc'le:. FIGUtRE 9. Coro w~ithout legunuse in a twxu-y ear rotationl, yijelded 5.'1 bushel.; per acre. (Plot 5, Manure-Nitrate-Vetch Experiment. Photographed July Compare with Figure 8. See Table 3. 11, 1926.) The Value of Legumes in Continuous Cropping Systems.Ih)ring the last fixve years. 1925 to 1929 iinclusive. exp~eriments were Con(Iuctedl at the Alabama Experiment Station to studyv the effect of v etch turnled at dlifferent dlater on the yields of cotton Vetc h was tulfled on a pproximnately the folloxwing ad 111Corn. dlates: Pdarch 25, Apil 5. and April 15. Cotton and coin Wxere uisuially pl anted'( albiout ten d axs oi twoiixweeks{ after the vetc h was tur n ed. The corn ex perime nt, the resulIts ot wxh ich are shoxxn in Ta ble 4, i ncl uded pilots th at were fertilized with nitrate of soda at VIG;URE 10.- hlairyv etch which proluced 5,8811 pounds of green material pcr acre. (Plot 12, Time of Turning Vetch for Coin IExperiment. PhotoSee Table 4. graphed and samipled April 13, 1912(.) dlifferent rates. Txwo plots adjacent to each vetch plot xxere llanted on the same date as the retch plot. The rate of application of nitrate of soda wxas increased as the date of turning the vetch xvas dlelayed1 because of the increasing amount of nitrogen app1lied1 by the late-turned retch. The experiment xvas (ixvided intoi txxo sections for spacing tests, one section being sp~aced1 18 inches and the other 36 inches in the drill. Each pilot in the experiment wxas fertilized alike xWith phosphate and piotash. The average yields of corn folloxwing xvetch turned M\arch 25, April 5, and April 15 xvere 22.6, 29.3, and 29.5 bushels per TABLE 4.-Yields of Corn Following Vetch Turned at Different Stages ot Growth. Date of turning vetch III I Plot Fertilizer treatment.* Pounds per acre I Planting date Four-year average*** yield of legumes. Pounds green matter per acre i ! Average yields 1925**-1929, bus. Spacing 18" I Increase due to treatment, bus. Spacing 36" 18" I j I 36" 7.1 April 8.9 13.0 April 6.8 13.9 4.2 15.2 12.3 19.4 April 6.4 19.6 15.5 5,259 22.6 March 25 April 5 10.8 22.1 12.1 20.9 15.0 April 20 soda 21.7 25.4 April 20 18.3 12.9 soda April 5 29.3 24.9 8,849 22.2 16.1 April 20 22.1 19.4 15.0 10.6 May soda 19.5 26.6 22.3 13.5 May soda 12,072 25.8 22.4 29.5 17.0 . n I May r r I April 15 I rI * All plots received 400 pounds of superphosphate and 50 pounds of muriate of potash per acre in 1925 and 1926. One ton of 16 per cent basic slag per acre was applied to Plots 4, 8, and 12 in the fall of 1926, and on all other plots in the spring of 1927. No fertilizer has been applied since the spring of 1927. Hereafter, all plots will receive 1 ton of basic slag and 250 pounds of muriate of potash per acre once every 5 years. ** Yields in 1925 were very low due to drought. *** Vetch was killed by cold in January 1928. None 100 Nitrate of 200 Nitrate of Vetch *** 200 Nitrate of 300 Nitrate of Vetch *** 200 Nitrate of 400 Nitrate of Vetch *** soda soda 1, 5, 9 and 13 2 3 4 6 7 8 10 11 12 ,,, I 15 Fi(I ll 1~. Cor(n followi ng ea1v tn1-IP %(1 ch p n e( "6. n; acre. ( Plot 4, Time1 of iTurning th I'mt Con L\1 ): riflli giuj1)IhI July 11, 1926;.) Sue Table 4. aI I (LI . lPhito- aicre, resp~ectiv'ely. The piot on -which retch wvas tui~nedl under made a higher average yield in each case th an the adjacent nitrate lplots that weie pilanteu at the same time. Plots 3, 6, and1 1 0 receiyed 200 pounds of nitrate of soda per acre, but were lantedl at different dates. There was a v ariation of onlyv 2.7 bushels per acre in the yields on plots planted April 5, April 20, and May 1. Vetch xvas killed by cold in 1928 and the yields shown in Table 4 relpresent a comparison of four crops of vetch and fiv e aplplications of nitrate of soda. The average yield1s (If corn on Untreated1 plots were higher on the 36 inch spacing section, wxhereas, (In the treated1 plots the yields were higher when spaced 18 inches ap~art. The average results of the cotton experiment alre presented in Table 5. P'lot 3 wvhich was turned M\arch 25 prodluced 59 po(unds of seed cottoln pmer acre more than Plot 4 xwhich wvas tuirnedl April 5, and 133 poIundls more than Plot 6 which wxas turnedl April 15. These data indicate that early planting of co~ttoIn is a more important factor in this experiment than the increase in the growth of vetch on the later turned plots. Nitrate (If sod~a at the rate of 300) pou nds per acre was appllied on lull, 1 I, 1!2(. ('I l l I - k ith FigI . 11. see 'IUIIcI 1. Plots 2 and 7 ol this ex ioiinte nt. Plot 2 which Wast pllantedl on April 5) madle 49 p~ounds of seedl cotton p)er acre mlor'e than Plot 7, which was planted on April :;0. 'This further empilhasiz/es the In other w ords, turning imrn~tance of planting cotton early. griow\th has lbeen made to supp)Ily legume',s as Carlya sufiit a reasonable anounilt ot nitrogen is p~referale to declay ing turnig until maximum griowth has heen madle and subseqIuently delaving the planiting (late of cotton. The data ini Table 5 showe that appllications of 800 p)ounds of nitrate of sodla per acre produced 68 p0)u11(1 ot seed cotton per acre mre than early tunned v etch and 152 pond1(s mfore2 than The planting (lates of the corresponding late turnedl vetch. 11e inicreases iln yields5 vetch and niitrate 1)1ots were the same. of seed cotton per' acre (lue to vletch ranged fromn 5)9: pounds for early tuned to -460 pounds for late tuin1eul vetch. All plots in this experiment Welrc oniformly treated wxith phosphorus and ptotash. 18 TABLE 5.-The Effect of Vetch Turned Under at Different Stages of Growth on the Yields of Cotton. Plot Legume and nitrate treatment* Date of turning vetch Date of nlanting cotton FiveFiveyear year Average average average increase over yields of yields of green seed check vetch cotton plots per acre per acre Pounds 5 & 8 None 2 Nitrate of soda March 25 3 Vetch 4 Vetch April 5 April 15 Vetch 6 7 Nitrate of soda April 5 April 5 April 5 April 20 April 30 April 30 4,542 9,757 9,992 Pounds 373 1,034 966 907 833 985 Pounds 661 593 534 460 612 * Each plot in the experiment received 600 pounds of superphosphate and 75 pounds of muriate of potash per acre annually. Nitrate of soda was applied to Plots 2 and 7 at the rate of 300 pounds per acre. RESIDUAL EFFECT OF LEGUMES The Effect of Vetch and Annual Yellow Melilotus Turned in 1925, 1926, and 1927 on the Yields of Corn and Cotton in 1928. -The severe freeze in January 1928 killed monantha vetch and annual yellow melilotus on two soil fertility experiments. Immediately after the freeze the land on which these experiments were conducted was plowed to destroy the scattered plants that were not killed by cold. Vetch was grown on the land the three years preceding the freeze in the experiment comparing vetch turned at different dates for corn. The dates of turning vetch, planting corn, and the amounts of nitrate of soda applied to adjoining plots are shown in Table 4. The average yields of green material turned under prior to 1928 and the yields of corn in 1928 are shown in Table 6. The data presented in Table 6 show that the vetch residue on the plot which was turned early produced more corn than 100 pounds of nitrate of soda on an adjoining plot. The plot on which vetch was turned April 5 produced more corn than adjoining plots which received 200 and 300 pounds of nitrate of soda per acre. The vetch residue on the plot which was turned late produced nearly twice as much corn as 200 pounds of nitrate of soda and approximately as much as 400 pounds of nitrate of soda on adjacent plots. The comparisons are made on the section in which corn was spaced 18 inches apart. TABLE 6.-Residual Effect of 1925-1927 Turned Vetch on 1928 Corn Yields. Pounds of nitrate of Date of Date of vetch. Ibs. 3 yr.* av. yield of green wt. per acre 1925-1927 Yields of corn in 1928. Bushels per acre Spacing 18" 36" Plot turning vetch planting corn soda per acre Increase over 1928 check yields. Bushels per acre Spacing 18" 36" 1 & 5 2 3 4 5 & 9 6 7 8 9 & 13 10 11 12 March 25 April 5 April 15 April 5 April 5 April 5 April 5 April 5 April 20 April 20 April 20 April 5 May 1 May 1 May 1 100 200 4,629 200 300 8,147 200 400 12,014 19 6.1 11.7 22.6 16.0 9.4 22.0 26.3 28.9 12.9 25.0 36.8 34.6 7.3 12.1 14.7 13.9 8.3 15.7 16.9 17.6 7.9 21.7 29.0 22.8 5.1 16.5 9.9 12.6 16.9 19.5 12.1 23.9 21.7 4.8 7.4 6.6 7.6 8.6 9.3 13.8 21.1 15.9 * Monantha vetch on these plots was killed by cold in 1928. 20 In another experiment vetch and annual yellow melilotus, with two rates of liming, were grown the three years preceding the freeze. The legumes were turned under each spring and followed by cotton. The fertilizer treatment and the average yields of cotton are shown in Table 10. The average yields of green material produced by the legumes in 1925 and 1927 and the cotton yields in 1928 are shown in Table 7. The yields of green material for 1926 were not recorded. Melilotus practically failed in 1925. TABLE 7.-Residual Effect of Vetch and Annual Yellow Melilotus Turned in 1925, 1926, and 1927 on 1928 Cotton, Yields. Average yield Plot Legume crop green Pounds of seed cotton per acre Lbs. per acre 1925 & 1927 matter. 1928 Increase over yield of check plots 1, 4, 7 and 10 2 5 6 8 9 None Vetch Vetch Melilotus Vetch Melilotus 8,983 10,713 10,010* 11,393 4,810* 103 204 348 308 426 348 101 245 205 323 245 * 1927 yield only. A study of the 1928 cotton yields presented in Table 7 shows that all plots on which legumes had grown produced more cotton than the check plots. The average increase of the legume plots over the check plots was 224 pounds of seed cotton per acre. The residual effect of legumes shown in Tables 6 and 7 would probably have been greater in 1928 but for the fact that heavy rains in late May and early June of that year leached some of the nitrogen out of the soil. As evidence that leaching took place, analyses of soil samples taken at different depths on July 21, 1928 showed that 73 per cent of the nitrogen found was from a layer 16 to 32 inches below the surface of the soil. These analyses also showed that 38 per cent was found at a depth of 24 to 32 inches. Further evidence of leaching was indicated by the appearance of the plants on the plots which received no nitrogen and on the legume plots which showed no visible difference in color or size before the last of June. On Plot 9 of the experiment comparing manure, nitrate of soda, and vetch as sources of nitrogen for cotton and corn, monantha vetch was killed by cold in 1928. In 1929 monantha vetch and Austrian winter peas on this plot were so severely damaged by rabbits that only 809 pounds of green matter per acre were produced. According to the analyses made in the laboratory this green matter contained 5.2 pounds of nitrogen. By comparing the yields of the unfertilized check plot with plot 8, which received 325 pounds of nitrate of soda, it was found that 21 one pound of nitrogen produced 0.53 bushel of corn. Calculated on this basis the 5.2 pounds of nitrogen contained in the vetch and peas turned under in 1929 would have produced 2.8 bushels of corn per acre. Plot 9 produced 15.9 bushels more corn per acre in 1929 than the check plots. Since 2.8 bushels of this difference could have been due to the legumes turned under in 1929, the residual value of 1927 turned vetch to the 1929 crop of corn may be assumed as 15.9 less 2.8 or 13.1 bushels per acre. Effect of Kudzu on the Yields of Succeeding Crops.-In another experiment kudzu was planted in the early spring of 1916 and turned under in the spring of 1919. The kudzu made little growth in 1916, covered the ground in 1917, and made a dense growth in 1918. From 1919 through 1929 two crops of sorghum hay, four crops of corn, and seven crops of oats were grown on the area devoted to this experiment. The average yields of sorghum hay, corn, and oats are presented in Table 8. TABLE 8.-Influence of Kudzu on the Yields of Succeeding Crops. Average yields of following crops Plot 7 Crops oats. Growthof 2 Crops sorghum 4 Crops corn. kudzu hay. Bushels per acre Bushels per acre Pounds per acre No kudzu Kudzu 3,264 5,800 14.7 34.0 16.6 24.5 1 2 The results in Table 8 show that the residue from kudzu produced an average increase of 2,536 pounds of sorghum hay per acre in 1919 and 1920. The average yield of four crops of corn following kudzu was more than double the yield on the plot that had not grown kudzu. The average yield of seven crops of oats on the kudzu plot was 7.9 bushels per acre more than that on the plot which had grown no kudzu. In 1929, ten years after the kudzu was turned under, the kudzu plot produced 9.2 bushels of oats per acre more than the plot on which kudzu had never grown. CULTURAL EXPERIMENTS WITH LEGUMES Effect of Fertilizing Materials on the Growth of Legumes.In order to determine the influence of superphosphate; superphosphate and ground limestone; and superphosphate, ground limestone, and stable manure on the yields of winter legumes grown in rotation, cuttings were made from plots so fertilized in rotation experiments located near Andalusia, Hackleburg, Prattville, and Sylacauga. The yields reported in Table 9 were calculated on the basis of cuttings of the green tops from two or three small areas on one-twentieth acre plots. The areas harvested totaled approximately 100 square feet. Monantha vetch was the crop harvested at Andalusia during each of the three r A / .4. aII' .. . 1'1 : . I I aiir xetcl( ,oa tulftrliIred plot (No. 5)o txO- VQ0 IitI(tii a exei meat on Norfoli k sa aix -oil nlear An ai a. Yiel~d, 120) ptoni n (((I coitton peti ((ci. See i gs. 11 andii 15. (Plhotogvaphed Apil 1, 192(;.) Sce Table 9. years; at Ilackleboug hairy vetch wvas used in 1927 and a mixture of hairy and monantha vetch in 1929 ; at Prattville the crop was monantha \'etch ;and at Sylacauga hairy retch was growni. The Superp hosphate andi stable manure wvere each appl ied1 imnmedliately l)Qfore( planting the winter legume. The groundi l im estone Wvas a ppllied1 in 1928 at HacklIeburg, Syl acauga, and P~rattv ille. and in 1921 at Andalusia. The rates of application per acre wvere as follow s: 400) pounds of sup~erphosp~hate, 4,0()0 TABLE 9.-Influence of Superphosphate, Ground Limestone, and Stable Manure on the Yields of Winter Legumes in Rotation. Grco% igchtpe Soil treatmenC~t And11(1ausia av.erage 3 crp II ace IrUIC average 2 crp Pra-ttv.ille 1929 c101) Sylacua 19127 cropi 1921-28-29 19127 K 1929 2,269 No treatment S up e rpo sp hate~ Su perphosp51hate. Gro'iund lim~estonie 4,.865 1:1,1 81 :3,6:30 5,929) 11,071 11,071 1,589r 10,802 5,4174 7,715 20,542 24,054 14,665 18.576 Sup erphousph ate Ground iii limes~toneC Stable mianure 16.798 ~~ kY 4 1 l'I(G I U' vetch fertilized with phosphatc (l'ot No. 2) in a two Yield, 330( pounds of seed cotton per acre. See Figs. 13 and 15. (Andalusia Field. Photographed April 1, 192(6) Sec Table 9. 4. llaiiy eiear rotation on Norfolk: soil near Andalusia. at Sylacauga), anl 6 tons of stable manure. The importance of superphosphate for winter legumes is emp~hasizedl by the results given in Table 9. The increase due to this material was 8,321 pounds of green matter per acre or aplroximately 20() per cent at Andalusia. At Hackleburg the increase wvas 8,533 lounls per acre with a still greater percentatge increase than at Andalusia. The single crops harvested at Prattville and Sylacauga both further emphasize the imlortance of p~ounds of groun1 limestone (marble (ust usedl lphosp1horus. The imlortance of lime under the conditions of these tests is almost as great as that of phosphate. The increases (due to lime were 7,356 p~ounds at Andalusia, 8,863 pounds at Hackleburg, and 5.142 p~ounds per acre at P'rattville. The slight loss at Sylacauga may be due to the possibility that enough marble dust particles to satisfy the lime requirement of the soil may have been blown from a marble refuse crushing plant located approximately a mile from the area. Six tons of stable manure per acre used in connection with sulperlphosphate and ground limestone increased the yield of 4 1. LAI - - -- - - 1( IAZ 1' 5, Hairy vetch on limed plot (Noo. ) ferctilied wxith p~hosphate. Yield, 1,050 piounds of seed cotton per acre. See F'igures 1:; and 14. (Anodalusia Field. Photographed April 1, 192~6.) See Table 9. green material per acre from 20,542) pounds to 24,054 pounids at Andalusia. from 1 4,665 pounds to 18,876 pounds at 1lackleburg, from 7,718 p)0und(1 to 16,798 1)0u11(1 at Sylacauga, and caused no increase on the single crop harvested at Pr'attville. In anlothler ('elimenlt conlductced by thet Alabiama Exp~erimenlt Station at Auburnll onl Norfolk sandl~y loam soil (luring the pa1st fiv e vears, x etch anld alnnua1l yellowx melilotus wvere grown dur ing the winter on 1)10ts that were cropped each year to cotton. Plots 5 and 6 of this experiment receiv ed all application of two tonls of grounid limestone peIr acre eveery five years. Plots 8 and 9 receixved 400) ploundS of' 16 11cr cent basic" slag ailnltallV- ins'tead( ol' superph~)losphlate as ai sourcie of p~hosp~horuls and an add~ition~al appllicaItion of olle-halt tonl of" basic slag 1per acre ever 'v fixe earlls. Oil Plot 3, wxhich had 1n( lime. anniual yellow melilotus failed each vealr. Although the pilan~ts on this plot xwere inoculated and suplpliedl with sufiient phosp~horus a11n( potash, none of them liv ed long enlough to make growth. The average resuilts of' this exp~erimenlt ill Table 10) shoxx that on Plots 6 anld 9 the additionl of' lime inc reasedl the growth of' mel ilot us, which whe(n1 turnedc~ 1111(1 rlIesulIted ill all incealse otf 378 poundtS of seed cot ton per acre on P~lot 6 and~ 312 pounds on Plot 9. These lesults show that Plot 6., which lreiv ed 1gromio limestone, produced 66 TABLE 10.-Effects of Vetch and Annual Yellow Melilotus With and Without Lime on the Yield of Cotton. Average 4 yields of legumes. Lbs. of green matter per acre Plot No.t Fertilizer treatment. Pounds per acre Legume crop Pounds of seed cotton per acre Five-year 1925-1929 Increase over average of check plots average 1, 4, 7, and 10 2 3 5 6 8 9 400 50 400 50 400 50 400 50 2 400 50 2 400 50 /2 400 50 1/2 Superphosphate Muriate of potash Superphosphate Muriate of potash Superphosphate Muriate of potash Superphosphate Muriate of potash Tons lime 1 Superphosphate Muriate of potash Tons lime 1 Basic slag Muriate of potash 2 Ton basic slag Basic slag Muriate of potash 2 Ton basic slag None Vetch Annual yellow melilotus 3 Vetch Annual yellow melilotus Vetch Annual yellow melilotus 7,418 None 8,157 260 728 237 794 468 -23 534 5,537 8,080 638 719 378 459 2,648 572 312 1 Lime every 5 years. 2 When Plots 5 and 6 were limed. 3 Melilotus failed on this plot due to a lack of lime. 4 Average of 3 crops of vetch and 2 crops of melilotus. 25 26 pounds of seed cotton per acre more than Plot 9, which received basic slag. The average results and other details of this experiment are presented in Table 10. The difference between cotton yields on vetch and melilotus plots that were limed was due largely to the fact that the stand of melilotus was reduced by freezing some years when vetch was not injured. A further study was made on the fields at Hackleburg, Sylacauga, Prattville, and also at Cusseta to determine the influence of ground limestone on the yields of summer legumes. The summer legume on which yields were taken followed an oat crop which was grown after corn in a regular three-year rotation. Either cowpeas or soybeans were used as the summer legume on these areas. Table 11 gives the average green weight from three limed and three unlimed plots on each area from 1923 to 1929. TABLE 11.-Effect of Lime on the Yield of Summer Legumes Following Oats in a Three Year Rotation. Rotation: Legume Winter Legume, Cotton, Winter Legume, Corn, Pounds green ma- Oats, Summer Percentage from liming Field Hackleburg Sylacauga Prattville Cusseta Average Soil type Ruston loam Decatur clay loam Greenville sandy loam Cecil clay loam teriincrease acre* terial per Unlimed** Limed** 5,139 3,911 4,957 3,315 4,331 8,172 5,259 8,293 5,569 6,823 59 34 67 68 58 * Average six crops at Hackleburg, five crops at Prattville, and four crops each at Sylacauga and Cusseta. ** Average of three plots. A study of Table 11 reveals the fact that lime increased the average yield of green material by 34 per cent on Decatur soil, 59 per cent on Ruston soil, 67 per cent on Greenville soil, and 68 per cent on Cecil soil. The average increase due to liming was 58 per cent. Time, Rate, and Methods of Seeding Winter Legumes.-An exerl)imnlt comp~arinig the effect of dIifferent methods ot seeding hairy v etch. monanthit \ etch, andI Austrian xxinter peas wxas started in the tall of 19 26 and completed in the spring of 1 929. Plantings werl madle iin drills 12 inches apart andl broadlcast Sep1tenmber :10. Oc tobler 26, November 28. and D~ecembei I19. Seeding wxas at the rateI of 20 loundls per acre for hairy and nmnarntha vetcht and~ 45 piounds per. acre for Austrian winter pecas. All 1lots in the experimnit were harvested at the same time andI the of giceti matter per acre wveie deteinedl~l. The ave(rage aitloltls yiclds and other dletails of the explerimlent ate showvn in Tfable 12. A st tdy of the results presentedl in Tfable 12 show-s that each of the three legumes nmade higher yields when drilled than wxhen The relative( increase ini jields duew to drtilling wxas brioad~cast. on the November and 1)ecemler lauiltings than on the grter The increase (due to drilling. howexver, does earlier dlantinigs. The thiree-year averaige' nt offset the effects of late pilanting. yild of hairy v etch wxas low xxhen dlrilledl and~ when hroadlcast. The unuua~lly loor growthl of hairy \ etch itl t his explerimntt wxas ' ~ At ~ f * /, ".9> . .4V~% ! v , . . 7d IGURE 16. Austrian winter peas shown in the foreground were seeded (ExperiNovember 23. September 29 seeding shown in the background. Photographed ment on Time and Rate of Seeding Winter Legumes. See Table 13. April 4, 1926.) TABLE 12.-Pounds of Green Material Per Acre on the Methods of Seeding Experiment of Hairy Vetch, Monantha Vetch, and Austrian Winter Peas 1927-1929. 28 29 due largely to the effect of anthracnose which seriously injured the 1928 and 1929 crops. November and December plantings of monantha vetch and Austrian winter peas failed to produce sufficient yields of green material to supply enough nitrogen for the needs of succeeding crops. Time and rate of seeding are important factors in the production of winter legumes. The results of many winter legume experiments prior to 1927 indicated that a large per cent of the failures with these crops had been due to late planting. An experiment was conducted during the three-year period, 1927-1929, to determine the effect of the date and rate of seeding hairy vetch, monantha vetch, and Austrian winter peas. The dates and rates of seeding and the average results of this experiment are presented in Table 13. TABLE 13.-Pounds of Green Material Per Acre on the Time and Rate of Seeding Experiment of Hairy Vetch, Monantha Vetch, and Austrian Winter Peas. Hairy Vetch Seeding rate. Pounds per acre Yield per acre. Pounds 1927-29 Monantha vetch Seeding rate. Pounds per acre Yield per acre. Pounds 1927-29 Austrian winter peas Seeding rate. Pounds per acre Yield Per acre. Pounds 1927-29 Planted September 30 10 20 30 10 20 30 10 20 30 1,027 1,577 2,228 791 1,682 1,893 354 676 918 10 20 30 10 20 30 8,393 9,240 9,211 5,308 6,949 7,700 30 45 60 30 45 60 30 45 60 7,412 8,138 8,653 4,747 5668 6,931 Planted October 26 Planted November 23 10 1,442 2,000 20 I 2,930 30 Planted December 19 I I I 2,313 3,322 3,669 10 20 30 None None None 10 20 30 594 947 1,312 30 45 60 835 1,393 1,562 A study of the data presented in Table 13 shows that the yields of hairy vetch were low on all dates and at all rates of seeding. The yield from 20 pounds of seed per acre more than doubled the yield secured from 10 pounds on the October planting. Twenty pounds of seed produced 70 per cent as much green matter as 30 pounds when planted in September. When planted in October 20-pounds of seed produced 83 per cent as much green matter as 30 pounds. November and December plantings practically failed at all rates of seeding. 30 Monantha vetch produced sufficient green matter to supply approximately 60 pounds of nitrogen per acre when seeded at the rate of 10 pounds per acre on September 30. The value of early planting is well shown by the fact that 10 pounds of seed planted September 30 produced more green matterthan 30 pounds planted October 26. There was no material difference in the yields secured from 20 and 30 pounds of seed. A satisfactory crop was not produced when seeded on or after November 23 regardless of the seeding rate used. Average yields in Table 13 show that 30 pounds of Austrian winter peas planted September 30 produced as much green matter as 60 pounds planted October 26. No rate of seeding used produced a satisfactory yield of green matter when planting was delayed until November 23. The results of this experiment prove conclusively that early planting is absolutely essential to success in growing the three legumes used in this study. These data also show that a material saving can be made in the amount of seed necessary for success if planting is done early. The advantage of early seeding is shown in Table 14 which gives the relative yields of vetches and related species grown in rows three feet apart on sandy upland soil at Auburn. The contrasted planting dates were October 1 and November 1. Only those kinds are considered that were harvested in five, four, or three years, as shown in other tables. TABLE 14.Increased Yield of Air-dry Matter from October over November Planting. Kind In entire season Pounds Per cent increase 1,810 337 -176 899 795 539 378 499 729 645 70 11 -5 37 32 21 17 14 32 25 To April 1 Pounds increase 1,148 1,024 564 860 769 535 551 1,450 599 833 Per cent 190 122 76 10 136 97 103 174 67 108 Monantha vetch Woolly-pod vetch Oregon vetch Austrian winter peas Hairy vetch Crimson clover Hungarian vetch Tangier peas Purple vetch Average for 9 species The average increases per acre in dry material due to early planting were 645 pounds or 25 per cent for the late harvest, and 833 pounds or 108 per cent for the early harvest. Growth and Yields of Winter Legumes.-The rate of growth of winter legumes is of vital importance. The earlier the legumes make sufficient growth to be turned under the earlier the crop that follows may be planted. The importance of early turning of vetch for cotton is shown by the cotton yields on Plots 3, 4, 31 and 6 in Table 4. An experiment was conducted at Auburn during a three-year period, 1927 to 1929, to determine the amount of growth that would be made by hairy vetch, monantha vetch, and Austrian winter peas planted in drills 12 inches apart at different dates in the fall and harvested at different dates in the spring. The planting dates, harvesting dates, and three-year average yields for this experiment are shown in Table 15. TABLE 15.-Pounds of Green Material Per Acre on the Rate of Growth Test of Hairy Vetch, Monantha Vetch, and Austrian Winter Peas. Date of Planting Dates of harvesting March 10'" March 22 April 2 April 17 1927-29 1927-29 1927-29 1927-29 Hairy vetch Sept. Oct. Nov. Dec. 30 26 23 19 2,581 1,569 263 82 2,311 1,633 415 119 2,377 1,792 630 160 2,023 2,136 883 274 1,539 2,072 .1,001 344 May 2 1927-29 Monantha vetch Sept. Oct. Nov. Dec. 30 26 23 19 5,635 3,245 1,021 255 7,400 5,408 1,841 373 7,719 6,027 2,657 670 8,602 6,596 4,350 1,393 6,087 5,000 5,226 2,090 Austrian winter peas Sept. Oct. Nov. Dec. 30 26 23 19 4,728 2,846 978 319 5,909 4,265 1,545 426 6,996 5,948 2,372 761 9,436 8,724 4,548 1,756 7,515 7,116 4,745 2,446 * Harvested March 19, 1929. The results in Table 15 show that monantha vetch produced more green matter than either hairy vetch or Austrian winter peas on all cuttings up to April 2 except on December plantings. Austrian winter peas produced the highest yield on all cuttings after April 2 except on the November 23 planting. The results of this experiment show that monantha vetch and Austrian winter peas made sufficient growth to turn under by March 22 while hairy vetch made slightly more than half as much growth by that time. Many kinds of winter legumes were planted in rows three feet apart for harvesting in the spring of 1925, 1926, 1927, 1928, and 1929. A part of each experimental area where the stand was practically perfect was cut close to the ground at various harvest dates each spring and the forage immediately weighed and again weighed after thorough air drying for a number of weeks. Table 16 records the acre yields at the stage of late full bloom from annual plantings made near October 1 and from those made near November 1. TABLE 16.-Average Yields in Entire Season of Winter Legumes Planted in Rows October 1 and November 1. Number of years Pounds per acre Planting date Oct. 1 1 Green Dry. Green Kind averaged Planting date Oct. 1 Nov. Nov. 1 Dry Monantha vetch Woolly-pod vetch Oregon vetch Austrian winter peas Hairy vetch Crimson clover Hungarian vetch Tangier peas Scotch vetch Subterranean clover Purple vetch Canada field peas Narrow-leaved vetch, Southern Red clover, first year 5 5 5 5 5 4 5 3 2 3 3 2 5 5 5 5 5 5 4 4 2 2 3 3 5 1 16,701 11,664 11,424 11,758 11,264 10,760 10,612 22,826 7,445 8,009 7,815 9,080 4,378 3,419 3,003 3,317 3,271 3,090 2,560 4,165 3,930 3,805 3,022 2,240 8,168 12,220 10,416 8,392 9,560 6,200 8,070 10,790 11,169 4,880 7,573 6,200 6,192 3,220 2,536 3,082 3,179 2,418 2,476 2,551 2,182 3,666 3,293 2,656 2,293 2,744 1,467 1,033 33 Monantha vetch ranked first for the entire season in average dry weight of the October and November plantings, with woollypod vetch second and Oregon vetch third. Next came Austrian winter peas, hairy vetch, and crimson clover, all with practically identical average yields of air-dry material. Austrian winter peas when allowed to stand until late bloom stage were sometimes reduced in relative rank by the attacks of plant lice. This injury was especially severe following the mild winter of 1926-1927 and again in 1929 when additional injury to this legume was caused by diseases that resulted in a blackening of the leaves and the stems. Tangier peas ranked at or near the top in years of mild winters but were completely killed in two out of five years. Likewise frequent winter killing subtracted from the apparent productiveness of subterranean clover, purple vetch, and Canada field peas. Earliness as well as total yield is important, especially when winter legumes are to be used for soil improvement. Therefore, in Table 17 these plants grown in rows three feet apart may be compared on the basis of the yields made by each on or just before April 1. Among the legumes grown in rows in all five years woollypod vetch led in yield produced by April 1 with 1,352 pounds as the average weight of air-dry matter per acre. It was followed by Austrian winter peas with an average of 1,284 pounds and by monantha vetch with 1,179 pounds of air-dry material. The average of October and November plantings was 1,023 for Oregon vetch, 931 for hairy vetch, 818 for crimson clover, 805 for Hungarian vetch, and only 331 pounds for the southern strain of narrow-leaved vetch. Table 17 shows that Tangier peas, among the kinds which were occasionally winter killed, led the list in yields when it survived, producing an average of 1,556 pounds of air-dry matter per acre. TABLE 17.-Average Early Yields (by or just before April 1) of Winter Legumes Planted in Three-Foot Rows October 1 and November 1. Number of years averaged Planting date Oct. 1 Nov. 1 Green Kind Pounds per acre Planting date Oct. 1 Dry Green Nov. 1 Dry Woolly-pod vetch Austrian winter peas Monantha vetch Oregon vetch Hairy vetch Crimson clover Hungarian vetch Tangier peas Purple vetch Canada field peas Narrow-leaved vetch, Southern 5 5 4 5 5 4 5 3 3 2 5 5 5 5 5 5 5 4 4 3 3 4 9,280 9,136 10,270 7,440 6,546 6,520 6,027 13,987 7,387 2,960 2,416 34 1,864 1,714 1,753 1,306 1,336 1,086 1,081 2,281 1,490 839 461 5,215 4,888 3,715 3,528 3,142 3,460 3,045 7,125 4,080 3,333 869 840 854 605 742 567 551 530 831 891 863 200 35 Relative Earliness of Winter Legumes.-Earliness is a desirable characteristic in any winter legume whether it is grown for soil improvement or pasturage. Some measure of the relative earliness of different species is indicated by the dates at which they bloom. The most definite date in the blooming period, and probably the most significant as a measure of relative earliness, is the date when blooming begins. Hence, for each of the crop years 1925 to 1929 inclusive, systematic records have been made, showing the dates of earliest blooming for each of the best known winter legumes and for a number of rarely-grown kinds. These dates have been recorded for plants from plantings made at least four times a year, namely, near the first of October, November, December, and the latter part of February. For reasons of space, the only data summarized in Table 18 are those relative to the initial blooming dates of vetches, peas, vetchlings, and crimson clover, and for only the annual November planting of each. The species are placed in the table in order of earliness. TABLE 18.-Relative Earliness of Winter Legumes by Initial Blooming Dates, from Plantings Made near November 1. Initial bloom dates Kindvetch Av. '26, '27, '29 Scotch vetch Narrow-leaved vetch, southern Bitter vetch Monantha vetch Crimson clover Woolly-pod vetch, commercial Tangier peas Oregon vetch Hairy vetch Austrian winter peas Purple vetch Days earlier than hairy Av. '26, '27, '29 28 Av. '27, '28, '29 Kind Av. '27, '28, '29 Mar. 24 Apr. Apr. Apr. Apr. Apr. Apr. May 1 4 8 13 16 21 8 Apr. Apr. Apr. Apr. Apr. Apr. Apr. 2 4 5 12 14 18 22 20 17 13 8 5 - 20 18 17 10 8 - 4 - Apr. 25 _-17 Later (Later) -3 As compared with the initial blooming date of hairy vetch, Scotch vetch was earlier by 28 days, the acclimatized strain of narrow-leaved vetch by 20 days, bitter vetch by 18 days, monantha vetch by 17 days, crimson clover by 10 to 13 days, the commercial strain of woolly-pod vetch by 8 days, Tangier peas by 5 days, and Oregon vetch by 4 days. Austrian winter peas were 3 days later than hairy vetch and purple vetch was 17 days later than hairy vetch. In general the order of earliness from the other three plantings was the same as for that made in November. 36 NITROGEN CONTENT OF WINTER LEGUMES Per Cent of Nitrogen in Green Matter.-More than a hundred cuttings of hairy vetch, monantha vetch, and Austrian winter peas were made at various stages of growth to determine the amount of green matter per acre. Samples were taken from these cuttings and analyses made in the laboratory to determine the per cent of nitrogen in the green matter of each of the three legumes. Data showing the average per cent of nitrogen are presented in Table 19. TABLE 19.-Per Cent of Nitrogen in Green Matter of Hairy Vetch, Monantha Vetch, and Austrian Winter Peas.* Variety Number samples Per cent nitrogen 1.00 168 Hairy vetch 0.76 161 Monantha vetch 0.79 180 Austrian winter peas *Some of the samples analyzed were supplied by the agricultural department of the Central of Georgia railroad. Based on the average per cent of nitrogen for each of the three legumes shown in Table 19, it is safe to recommend that hairy vetch be turned under when it has made sufficient growth to cut 5,000 pounds of green matter per acre and that monantha vetch and Austrian winter peas be turned under when they will cut 6,000 pounds of green matter per acre. Stated in terms of green matter per hundred square feet, it is recommended that hairy vetch be turned when it will cut 12 pounds of green material and monantha vetch and Austrian winter peas when they will cut 14 pounds per hundred square feet. If the three legumes are allowed to make the growth outlined above, about 45 pounds of nitrogen per acre will be added to the soil. THE PRODUCTION AND HARVESTING OF WINTER LEGUME SEED Regular Variety Test.-During the past five years, 1925 to 1929 inclusive, variety tests of vetches and winter peas were conducted. The most promising varieties were included in these tests. One-half of the area devoted to each variety was cut for hay and the other half harvested for seed. Of the varieties tested, monantha vetch, Austrian winter peas, and Tangier peas were the only varieties that showed any promise as seed producers for this section. The average yields of seed for five years of the three leading varieties were as follows: Monantha vetch 139 pounds, Austrian winter peas 216 pounds, and Tangier peas 225 pounds per acre. Seed from the variety test was not harvested in 1928 because of excessive rainfall at harvest time. A good crop of seed was produced and if harvested would probably have increased the average yields of monantha vetch and Aus- 37 trian winter peas. hence the average yield of this variety would not have been Tangier peas were killed by cold in 1928;. affected. Although Tangier peas have produced some good crops of seed, this variety is not winter hardy, having been killed by cold in January 1928 and in December 1929. General Field Planting.-General field plantings of about five acres of monantha vetch and Austrian winter peas were made in 1928 and 1929 for seed production. The seed crop from these plantings was grown and harvested under average farm conditions. Excessive rainfall delayed harvesting in 1928, but approximately 400 pounds of seed per acre were harvested. Harvesting methods were studied in connection with this general field planting. The most satisfactory method used was to allow the plants to become dry, rake with a hay rake without mowing, and thresh in the field. Threshing was done with a grain thresher from which one concave was removed. The speed of the machine was reduced to prevent splitting the seed. Spring Planting.-Another experiment was started in 1928 to study the effects of spring planting, with different fertilizer treatments, on the yields of seed of monantha vetch and Austrian winter peas. Plantings were made about February 15. All plots in this experiment received a uniform application of 600 pounds of basic slag per acre. One plot planted to each legume was treated with 5 tons of manure per acre and another received 200 pounds of nitrate of soda per acre. The yields of seed produced in 1928 and 1929 from spring plantings for seed production were lower than the yields from fall plantings. COLD, INSECT, AND DISEASE INJURY TO WINTER LEGUMES Winter Injury.-Detailed notes were made of the effects produced on each of a large number of winter legumes by the coldest weather of each of the past six winters, 1924-25 through 192930. The severity and date of the coldest weather, according to the readings of the Weather Bureau thermometers in a standard weather shelter at Auburn, in each of these years is shown in Table 20. TABLE 20.-Minimum Official Temperatures at Auburn. Winter of 1924-25 1925-26 1926-27 1927-28 1928-29 1929-30 do. Lowest reading F. 22 9 11 8 24 16 Date Dec. 26 Dec. 28 Jan. 16 Jan. 2 Dec. 9 Jan. 19 Next lowest reading F. 24 21 24 13 27 18 19 Date Nov. Jan. Dec. Dec. Jan. Dec. Dec. 25 14 16 9 12 3 20 38 Records were made of the effects of these and other temperatures nearly as low on the seedlings of a long list of species, each grown from plantings made near the first day of October, November, and December, respectively. In Table 21 only the effects on the plantings of early October and November are summarized, since the direct cold injury to the December planting was sometimes complicated with fatalities due to the lifting or heaving of the soil. Not every species was subjected to all the cold periods mentioned above. The species that escaped with no material injury to the seedlings of the plantings made in the early days of.October and November were: hairy, woolly-pod and Hungarian vetches, the Southern strain of narrow-leaved vetch (Vicia angustifolia), and Austrian winter peas. It now seems safe to class the above mentioned kinds as winter hardy, at least as far north as Auburn, Alabama, (latitude 32/2 and with an altitude of about 750 feet) when the seeds are planted not later than the first week in November. The list below embraces the nearly hardy kinds under the conditions of these experiments. These species were severely injured or killed only by the repeated severe freezes of January 1-6, 1928, with a minimum of 8°F. in the Auburn weather shelter. They were: Oregon vetch, crimson clover, "white blooming crimson clover", red clover, alsike clover, white clover, black medic, Tifton bur clover, Hubam, and yellow biennial melilotus. The kinds just mentioned proved winter hardy against a minimum of 9°F. on December 28, 1925, and of 11°F. on January 16, 1927; hence may be regarded as winter hardy at Auburn except in unusually severe winters. Biennial white melilotus (sweet clover) seems also to belong in this class. The following plants proved tenderer than those mentioned above: monantha, bitter, and Scotch vetches, lentil, Tangier peas, sweet peas, and the northern strain of narrow-leaved vetch. These were materially injured at Auburn by a minimum temperature of 90F. or 11°F., or by both, and were sometimes notably injured by slightly higher temperatures. The kinds that were materially damaged (killed or severely injured) by the rather moderate cold of December 1929 (minima 18°F. and 19°F., with a thin coat of ice for one day over all foliage) were the following: monantha, purple and Scotch vetches, only the northern strain of narrow-leaved vetch, Tangier peas, Canada field peas, lentil, horsebean, sweet peas, hairy Peruvian alfalfa, California bur clover, and Fenugreek. The most tender toward cold of any of the winter legumes tested during the past six years were the following: horsebean, Canada field peas, purple vetch, Scotch vetch, sickle vetch, Lathyrus clymenum, Italian melilotus, early yellow annual melilotus, Egyptian or Alexandrian clover, early grass pea (Lathyrus sativus), late grass pea, chick pea (Cicer arietenum), Medicago TABLE 21.--Effects of minimum Winter Temperatures (Fahrenheit) on Winter Legumes. Kind Killed byF by Severely injured byF. Appreciably injured by F. Escaped material injury by F. Vetch group (Vicia, Pisum, and Lathyrus) Monantha vetch (V. monantha) do. Oregon vetch (V. sativa) do. Purple vetch (V. atropurpurea) do. Bitter vetch (V. ervilia) Narrow-leaved vetch, northern (V. angustifolia) do. do. Scotch vetch (V. sativa, var.) do. Taingier peas (L. tingitanus) do. Canada field peas (P. arvense) do. Pearl vetch Sickle vetch (V. sicula) Lentil (Ervum lens) do. Horsebean (V. faba) do. do. Sweet peas, Spencer mixed (L. ordoratus) do. Grass pea, Late white (L. sativus) do. Grass pea, Early white do. 8 19 8 18 9 18 8 11 24 9 11 22 24 24 22 24 24 9 24 11 8 11 9 18 11 11 22 11 11 19 11 9 11 11 24 1 8 19 8 19 8 19 8 18 8 9 8 9 8 9 18 8 18 8 9 8 39 9 9 18 9 18 24 11 24 24 TABLE 21.--Effects of Minimum Winter Temperatures (Fahrenheit) on Winter Legumes (Cont'd.). Kind Killed b byF. Severely injured byF. Appreciably injured byF. Escaped material injury byF. Clovers (Trifolium sp.) Crimson clover Subterranean clover do. Hop clover (T. aureum) (Cultivated) (In old sod) do. Medics (Medicago sp.) Alfalfa, common Alfalfa, hairy Peruvian Black medic (M. lupulina) Bur clover, southern (M. arabica) Bur clover, early southern Tifton bur clover (M. rigidula) California bur clover (M. hispida) do. Snail clover (M. scutellata) Button clover (M. orbicularis) Spineless bur clover (M. apiculata) M. hispida lappacea M. echinus 8 8 8 8 9 9 9 8 11 18 22 9 11 22 40 24 18 9 9 11 9 8 18 8 8 8 8 19 9 9 9 11 11 9 24 41 hispida lappacea, Medicago echinus, snail clover, and Lupinus manus. Variable Response of the Same Species to Low Temperatures. -It is not correct to assume that one species will always be killed by a definite minimum temperature. Other conditions not fully understood sometimes modify the severity of the injury. For example, monantha, purple and Scotch vetches, and Tangier peas withstood a minimum of 11°F. on January 16, 1927; yet these same species were among those completely killed by the much more moderate cold of December 1929. In the latter period the lowest temperatures reached in the cold periods near the first of December and again just before Christmas were 18°F. and 19°F., respectively. Plausible explanations of the more disastrous effects of these last mentioned temperatures are to be found in: (1) the continued repetition of below-freezing morning temperatures for six successive days just before Christmas and (2) in the maintenance of a thin coating of ice over all foliage for about 24 hours on December 22, 1929, this glaze being maintained while the thermometer hovered around the freezing point throughout the day. The following example shows that even during the same period of cold a single species may suffer various degrees of injury on different fields of the same farm. Monantha vetch, planted in rows three feet apart near the first of October and also near the first of November, was killed to the last plant by the December 1929 freezes, in the South Annex field at Auburn. But this cold weather killed completely only a small percentage of monantha plants sown broadcast in October of the same year on another Auburn field. In still other parts of the station farm monantha vetch plants of about the same age suffered intermediate degrees of injury. The stand that survived the two cold periods which occurred in December 1929 varied on this one farm from 0 to about 90 per cent in various fields having about the same slope or exposure. Possible accessory causes seem to lie in the unlike amounts of disease noted in the several fields and in the relative fertility of the various soils. Damping off had been especially severe in the South Annex field, but no special complaint of it was made as affecting serious losses in the other fields although it was prevalent. The red root rot disease described in later paragraphs was shown by repeated examinations and counts to be extremely prevalent on monantha vetch in the field where complete winter killing occurred. It was also quite general in the field where the next largest amount of winter injury was found and was least prevalent in the fields where the least amount of cold injury was noted. In one field about 90 per cent of a stand survived the winter of 1929-1930. Moreover, soil conditions other than the presence of disease organisms seemed to influence the severity of winter injury. 42 Relative Susceptibility to Insect Injury and Disease.-Plant lice or aphids have been found on all vetches tested. The extent of their injury has been very unlike on different species. The kinds most severely damaged by aphids have been monantha vetch, bitter vetch, and Austrian winter peas. Occasionally the growth of monantha vetch and Austrian winter peas has been stopped and the plants practically killed by plant lice. This injury has usually become severe only after April 1 when considerable growth has already been made. Thus the fertilizer value of these legumes for plowing under early has not been greatly diminished, although there has resulted a notable reduction for the entire growing season in the potential production of nitrogen, organic matter, and forage. Hairy and woolly-pod vetches have been seriously injured by anthracnose and another disease (Ascochyta), both affecting the stems and leaves. The stems, especially near the ground, turn black, and shrivel, and many of the branches die. The foliage becomes discolored and often drops. Austrian winter peas, and Canada field peas have sometimes lost a considerable percentage of their leaflets by such blackening. Narrow-leaved vetch has sometimes been seriously injured by the diseases mentioned above. Hungarian vetch has been more free from these, as well as from other diseases mentioned later, than has any other of the many vetches tested. The damping-off fungi have sometimes killed a considerable percentage of the seedlings of vetches and other legumes when continuous wet weather has followed closely after germination. Especially notable have been its winter injuries to bitter, monantha, and hairy vetches, and to Austrian winter peas. Monantha and bitter vetches have been extensively attacked and seriously injured by a disease that exhibits symptoms not found on other vetches. This disease, here tentatively designated as red root rot, is characterized by bright red spots and streaks or patches on the whitened underground part of the stem, on the roots, and even on the germinating seeds of monantha and bitter vetches. Later the reddened area sometimes encircles the root or lower stem, thus causing the death or extreme dwarfing of the plant. Unthrifty plants sometimes show reddening and other unhealthy symptoms on at least a part of the foliage. Doctor J. L. Seal, pathologist of the Alabama Experiment Station, has found bacteria and in other cases fungi associated with these red spots on the underground part of the plant but has not yet made positive identification of this disease. The red spots on the seedlings are usually first observable at or just below the point of attachment of the old seed, as well as on the germinated seed itself just below the seedcoat. These facts suggest that the organism is probably carried by the seed. Moreover, our observations seem to indicate that this disease becomes more prevalent after vetch has been grown for several 43 years in the same field. This suggests the probable advisability of rotating soil-improving crops as well as practicing rotation for the main crops. This course seems entirely feasible in view of the fact that crimson clover and Hungarian vetch among the well tested legumes have for the past six years proved resistant to the diseases that have at times been extremely injurious to Austrian winter peas and hairy vetch growing in adjacent rows. SUMMARY Experiments with legumes were among the first studies undertaken by the Alabama Experiment Station. The results of this earlier work emphasized the need of phosphorus and the inocUlation of legumes by the proper bacteria. This bulletin records results of experiments with legumes by the Alabama Experiment Station from 1896 through 1929. The yield of corn was maintained by legumes on land cropped continuously to corn during a 34-year period. Corn yields declined more than 50 per cent on land cropped continuously to corn, without legumes, during a 34-year period, 1896 to 1929. Yields of cotton and corn in a three year rotation with legumes were four times as great as those in the same rotation without legumes. When applied alone in a three year rotation phosphate and potash did not increase the yield of corn and increased the average yield of seed cotton only 19 pounds per acre, but when they were applied in the rotation with legumes the yields were increased by 293 pounds of seed cotton and 11 bushels of corn per acre. Cotton following vetch produced 1,231 pounds of seed cotton per acre as compared with 1,222 pounds per acre and 1,471 pounds per acre on adjacent plots which were treated with 325 pounds of nitrate of soda per acre and 5 tons of manure per acre, respectively. Corn following legumes turned under March 25, April 5, and April 15 produced greater average yields over a 5-year period than corn on corresponding plots which rece4ved 200, 300, and 400 pounds of nitrate of soda per acre. Cotton following vetch turned March 25 produced a greater average yield than that following vetch turned April 5 or April 15. The residue from vetch turned in 1925, 1926, and 1927 produced more corn in 1928 than either 200 or 300 pounds of nitrate of soda per acre. The residue from vetch and annual yellow melilotus turned under in 1925, 1926, and 1927 resulted in an average increase of 224 pounds of seed cotton per acre in 1928. Vetch turned under in 1925, 1926, and 1927 resulted in an increase of 13.1 bushels of corn per acre in 1929. Kudzu turned under in 1919 increased the average yield of two crops of sorghum hay by 2,536 pounds per acre, four crops 44 of corn by 19.3 bushels per acre, and seven crops of oats by 7.9 bushels per acre. Applications of superphosphate in rotations at Andalusia, Hackleburg, Prattville, and Sylacauga resulted in increases of 2,000 to 8,000 pounds of green material per acre in the yields of winter legumes. Lime applied in rotations increased the yields of green material of winter legumes 7,356 pounds per acre at Andalusia, 3,863 pounds at Hackleburg, and 6,142 pounds at Prattville. Six tons of manure per acre applied in connection with superphosphate and lime in rotations, increased the yields of green material of winter legumes by 3,500 pounds to 9,000 pounds per acre at Andalusia, Hackleburg, and Sylacauga. Applications of ground limestone and basic slag produced approximately the same yields of vetch, but ground limestone produced more than twice as much annual yellow melilotus as basic slag. Annual yellow melilotus failed on Norfolk soil without lime. Lime in a three-year rotation increased the yields of summer legumes 34 per cent on Decatur soil, 59 per cent on Ruston soil, 67 per cent on Greenville soil, and 68 per cent on Cecil soil. Hairy vetch, monantha vetch, and Austrian winter peas produced larger yields when planted in drills 12 inches apart than when seeded broadcast. Light rates of seeding of monantha vetch and Austrian winter peas when planted September 30 produced more green matter than heavy rates of seeding planted October 26. Plantings of nine winter legumes made in drills near October 1 averaged 25 per cent more air-dry material than plantings made November 1. Monantha vetch and Austrian winter peas produced sufficient green material to turn under by March 22; by this date hairy vetch had produced approximately half as much growth. Monantha vetch ranked first among many winter legumes in total yield of dry material produced from plantings made about October 1 and November 1 in drills three feet apart. Next followed in order, for the five year period, woolly-pod vetch, Oregon vetch, and Austrian winter peas. Harvests made April 1 of many winter legumes show that woolly-pod vetch, Austrian winter peas, monaitha vetch, Oregon vetch, and hairy vetch, in the order named, produced the largest yield of dry material from early October and November plantings. The relative earliness of various winter legumes, based on the beginning of blooming, was found to be as follows: Scotch vetch, the southern strain of narrow-leaved vetch, bitter vetch, monantha vtech, crimson clover, wooly-pod vetch, Tangier peas, Oregon vetch, hairy vetch, Austrian winter peas, and purple vetch. Results of more than 100 analyses show that the green tops of hairy vetch contain 1.00 per cent nitrogen, monantha vetch 0.76 per cent, and Austrian winter peas 0.79 per cent. Based on the nitrogen content of the three legumes studied, 45 it is recommended that hairy vetch be turned when the green tops from 100 square feet weigh 12 pounds, and that Austrian winter peas and monantha vetch be allowed to attain a growth of 14 pounds. Monantha vetch, Austrian winter peas, and Tangier peas produced the largest yields of seed in the variety test of winter legumes but have not been consistent seed producers. Among the drilled winter legumes that proved hardy at Auburn against all low temperatures of the past six years were hairy, Hungarian, and woolly-pod vetches, Austrian winter peas, and the southern strain of narrow-leaved vetch. Killed or severely injured only once, with a minimum temperature of 8'F., were Oregon vetch; crimson, red, Alsike, and white clovers; black medic; Tifton bur clover; and hubam and yellow biennial melilotus. The same minimum temperatures that in some winters killed or severely injured a given winter legume brought little or no injury to the same species in other winters. Austrian winter peas and monantha and bitter vetches proved most susceptible at Auburn to injury by aphids. Stem and leaf diseases repeatedly damaged Austrian winter peas, hairy vetch, and woolly-pod vetch. Hungarian vetch and crimson clover were among the legumes that escaped injury by these diseases.