B4038 llllY I965 ¢ Q a a . '26‘ ‘t ¢. Q‘ . . .~ Summarnj Several very promising grasses have been intro- duced in the Southwestern United States but have not been used widely because of one or more limiting factors. Kleingrass, Panicum colomtum L., appears to be adapted to a wide range of soil and climatic conditions, but seed production is hampered by a severe seed-shattering characteristic. Pretoria 90 blue- stem (Dicanthium annulatum (Forssk) Stapf) makes very heavy forage yields but produces low quality seed and is not winter-hardy north of Central Texas. Other introduced bluestems have similar seed produc- tion and adaptation characteristics but generally are less productive than Pretoria 90. Studies conducted over a period of 5 years con- cerned with seed production characteristics and man- agement practices are reported. Flowering and maturity in the kleingrass in- florescence progresses from the top to the base for several days. Seed shattering may occur at the top of the inflorescence before blooming is completed at the base. Harvesting 5-10 days following the first signs of seed shattering resulted in maximum seed yields and seed quality. If most of the seed heads are produced at the same time, peak seed yields are limited to a very few days. Seed heads may be initiated over a long period, especially in the fall, resulting in a longer period in which satisfactory seed yields are obtained. Yields of 5O to more than l0() pounds of seed material in each of two harvests varying in purity from 15 to 60 percent are obtained. A fall seed crop of kleingrass was obtained 30 days after a hay harvest. Similar yields were obtained on stands not cut for hay. These results indicate that several seed crops might be obtained or that several FORAGE AND SEED PRODUCTION OF INTRODUCED BLUESTEM ACCESSIONS, COLLEGE STATION, harvests would be possible if the seed could; removed by agitation witho-ut cutting the plant' The introduced bluestems produce a spring _ crop varying in yield from a few pounds to more 125 pounds per acre. Quality‘ also varies from as as l0 percent purity to oyiier 40 percent p Regrowth following the spring seed harvest re the stage for floral development in early Au but floral development occurs poorly and sl_ Removal (clipping) of the vegetative growt I August results in regrowth which initiates flo within 30 days and seed maturity before frost. If? vegetative growth is not removed in August, fl’ initiation and maturity are frequently delayed ‘ after frost. Two seed crops annually are possible , this management practice. F Irrigation of the introduced bluestems is in dry years for fall seed production in the C0 Station area. However,‘ unnecessary irrigation g maturity and reduces seed yields. i Maximum seed yields of Pretoria 90 blu were obtained 2 to 5 days after the tips of app“, \ mately 50 percent of the heads would release ‘_ with a gentle pull with the fingers. a Several experimental bluestem accessions I been evaluated_for forage and seed production. S Pretoria 90 is outstanding in forage productio‘ was used as a basis for comparing the newer I’ None of the experimental material was superi Pretoria 90 in forage production. A few of the a sions produced higher yields of better quality but were unsatisfactory vegetatively. Thus, n0n' the lines show sufficient promise to warrant in ‘ 4v 1961-63 Forage productionl A _ P 1961 1962 ccess|on or . ercent variety Species 1961 1962 1963 ‘eavess Pounds Perqsm Pounds Percent‘ seed set seed set Pretoria 9O Dicanthium annulatum 3.4 1.5 2.6 27.1 9O 8 42 11.7 Medio Dicanthium caricosum .9 .6 1.5 24.5 81 10 16 7.2 Gordo Dicanthium annulatum .6 101 53 4080 Dicanthium ann-ulatum 3.0 2.0 2.1 30.2 98 19 36 8.2 6141 Bothriochloa intermedia 2.3 1.8 1.8 17.6 113 44 18 12.8 T-20069 Dicanthium sp. 3.5 2.0 1.4 31.5 105 ‘ 10 28 5.6 T-20299 Dicanthium sericeus 3.4 1.7 1.2 27.9 51 18 50 11.8 2660 Dicanthium caricosum 1.0 184 58 4081 Dicanthium annulatum .9 124 42 3025 Dicanthium aristatum .8 87 47 T-4931 Dicanthium annulatum 2.2 1.8 23.9 59 8.6 3965 Bothriochloa intermedia 1.1 .8 16.2 18 19.4 5398 Dicanthium annulatum .9 1.3 16.1 22 18.0 4099 Dicanthium annulatum .4 1.7 16.6 16 15.0 5297 Bothriochloa intermedia .3 1.2 16.1 10 12.4 55 x 485-2 Dicanthium annulaturn .7 1.0 14.0 20 14.9 Seed production’ ‘Yields from upland farm near College Station; 1961 was from second-year stand, 1962 and 1963 were from a test established in zSeed production data were taken from the forage plots after the final harvest and indicate fall production and not total seed pr“ potential. “Leaf percentage based on September 1962 forage harvest. ‘Percent seed set in 1962 was based on weight rather than number and therefore is low but indicates accession differences. ous GRASSES have been introduced into Texas l. e the late 1930s. Many of these have come jmmon usage in areas of adaptation. These King Ranch bluestem, buffelgrass, weeping , blue panicgrass, Sorghum almum and per- ‘thers. Still o-ther grasses have been shown to fpted but have not been established extensively of unavailability of seed or high seed costs. iay result from poor seed yields, low quality, ilty of harvest or other seed productio-n prob- T‘ Among the grasses which have shown some of adaptation but which are limited by seed tion or poor seed production characteristics, merous members of the introduced bluestems, illy members of the Dicanthizrm complex. Some include Medio, Pretoria 90, Angleton and bluestems. Kleingrass, Panicum coloratum, ther example of a promising grass which is i ed because of seed production characteristics. his publication reports research studies con- primarily with seed production problems of it ass, Pretoria 9O bluestem, Medio bluestem, 4080 bluestem and other bluestem acces- e I. A plant of kleingrass showing growth habit and re. Seed Production Characteristics of Some Introduced Worm-season Grosses Ethan C. Holt, Professor, Department of Soil and Crop Sciences Kleingrass (Panicum coloratum L.), shown in Figure 1, is described as a perennial warm-season bunchgrass introduced from Africa and adapted to a fairly wide range of soil and climatic conditions. The stems are slender and very leafy, reaching a height of 31/24 feet at maturity. Plants within the species vary from upright to stoloniferous in growth habit, from glabrous to- densely pubescent and from light to dark green to bluegreen in color. Kleingrass initiates growth early in the spring and remains green late in the fall. Seed are pro-duced on an open panicle and tend to drop from the panicle as they mature. Pretoria 9O bluestem (Dicanthium annulatum (Forsslc) Stapf) is a tall, warm-season, perennial b-unch- grass adapted to the heavier soils in Central and South Texas and the Gulf Co-ast. Plants are 4 to- 5 feet high at maturity. The stems are somewhat coarse with broad leaves which carry high on the stems (Figure 2). Extremely high yields of leafy forage are produced. The foliage is a characteristic grey-green color. Medio bluestem (Dicanthimn caricosum A. Camus) is described as a variant of Angletongrass with fine stems and good seed production. The plants are fine stemmed, leafy, dark green and reach a height of 24-30 inches at maturity. The plants spread rapidly by prostrate stems to fo-rm a dense, solid ground cover (Figure Medio is best adapted to clay soils or sandy so-ils with a shallow clay layer south of a line from Waco to San Angelo, Texas. Figure 2. A plot of Pretoria 90 bluestem showing texture and stem enlargement (booting) in August. TABLE l. SEED YIELD AND QUALITY OF KLEINGRASS, CO STATION, 1958-59 ,_ ‘Pounds of seed per acre Percent seed set Yea - r June August Total June August ‘g2: ‘I958 ‘I27 74 201 30 35 1959 so so 11o, 42 11 Kleingrass seed yields vary from year to?‘ depending o-n cultural treatments, harvest priic, and, possibly, age of stand. See-d yields in 1958 , , A _ __, 1959 are presented in Table 1. Satisfactory yi Figure 3. A solid stand of Media bluestem showing texture and W6“: Obtained- in 1958 but not in 1959- Beta“ ‘ stem enlargement (booting) in August. the seed shattering problem and indeterminate fl . ing, many seed are lost prior to- harvest, and 0t; Accession 4080 bluestem (Dicanthium annulattim are immature at the time of harvest. Ant (F o-rssk.) Stapf) is very similar to Pretoria 90 in growth (blooming) begins at the top of the inflorescence habit and foliage characteristics. on the tips of the branches of the inflorescence progresses toward the base of the inflorescence. _ quently, shattering may be noted at the extrem‘ a of the inflorescence while florets near the base in anthesis. Not all inflorescences develop at same time. Thus, some inflorescences may be ing at the time others are initiating anthesis. Be T of these seed producing characteristics, manag 1.95;. i and harvest practices were studied in an effort determine how and when maximum seed yields l. be obtained. The introduced bluestems were earlier classified as Aiidropogori species b-ut extensive studies in Okla- homal in recent years have led to their classification as Dicanthium, Bothriochloa and Capillipedium species. These classifications are utilized in this report. In the work at College Station, emphasis has been place-d on the Dicanthium group because of better leafiness and forage characteristics, but the Bothrioc/iloa species in general are more hardy. ‘rcelarier, R. P. and Harlan, ]. R. 1955. Studies on old world S 1 h d f y , A bluestems. Okla. Agri. Exp. Sta. Tech. Bul. T-58. amp_€s W€re aljvest? at _requent lntenf Harlan, Jack R., Celarier, R. P., Richardson, w. L., Brooks, 1959 starting at the llfSt lndleatlons of shattering; M- H» and M6111“, K- L- 1953- Studifis 011 0751 World blue the earliest inflorescence and continuing until stems II. Okla. Agri. Exp. Sta. Tech. Bul. T - . th d h d h tt d Th “i Harlan, Jack R., de TVet, J. M. ]., Richardson, W. L., and _e See a S _a erg ' €_ Study We‘? Conduit. wlth both the first seed crop ln late sprlng and 1 Chheda, H. R. 1961. Studies on Old World bluestems III. _ _ _ _ Okla. Agri. Exp. Sta. Tech. Bul. T-92. l early fall seed crop. The results 1n Figure 4 l, I50 - :11} Seed yield '20 I _x_‘ _x_} Percent purity 3 90 May-June ° ‘ 1 A 1- . . . 3 Fsggzlesmbefi‘ A E Figure 4. The influence of it“ a Vi. \ ,4 Q harvest on seed yield and purl 3 ,/ \ ‘g kleiiigrass, 1959. Initial sampling 11> I >~ vest) was on date of first t F's - .0 , i, 6O I T, M0 "June x 6O >, shattering. 1n ! x X '- c , ,_x-—-X 0' a I * z o. so - ,1 _ - so 3 August September \ a.’ V‘ x’ X k a. .xlfixqgx— QXQQ ' \ ‘k x__x_' x/ 5on0}? Xv ¢XO Xxx" o l l 1 1 1 1 l 1 1 1 1 l L l 1 l l l 1 1 l l l l l 1 l l l m 2O 25 3O I 5 ' IO I5 0 May-August June-September la critical period for see-d harvest in both the I and fall seed crop of kleingrass in 1959. Peak i, yields occurred approximately l0 days follow- i} e first apparent seed shattering on May 22. Q, y, there was a period ‘of more than 8 days in ‘harvest of mo-re than 100 pounds of seed 'al per acre was possible. The peak period in ‘ll occurred 12-14 days after first apparent seed ing. .A much shorter time existed in the fall ich harvest of more than 100 pounds of seed ,'a.l was possible. However, there was a long h approximately 20 days, in which yields aver- about 75 pounds per acre. Apparently the rate dilnaturity during this period approximated seed shattering. set ranged from 45 to 60 percent during a l‘ period in the spring. This perio-d coincided ithat in which yields amounted to 100 pounds gre per acre. The actual peak in seed quality days later than the peak in yield. Fall quality gw, averaging only about 15 percent with no ‘icant peak. fields in these studies were determined by cut- binding, drying and then threshing the seed. a much higher percentage of immature spikele-ts threshed seed would be obtained resulting in total yields but reduced purity. Field com- might be expected to» result in less total yield tter quality because less of the immature ma- would thresh while green. ned at frequent intervals again in 1960 (Figure pparently, sampling was started to-o- late in the " since yields dropped to less than 2O pounds “Acre 2 days after sampling started. The fall seed grhanged very slowly in yield and quality. How- harvest either at the time of first obvious shatter- or during the follo-wing l0 days would have _ ed in the best yield and quality. A second peak ld and slightly improved quality occurred about i l ‘i P S I bserved shattering. y eed purity and potential seed yields were de-l TABLE 2. KLEINGRASS SEED YIELDS WITH VARIOUS MANAGEMENT AND FERTILITY PRACTICES, AGRONOMY FARM, I962 Fertilizer Clipping height,‘ (Ileefled seed Per eere. Pevflde treetment inehee 9/21 9/25 ‘I0/2 io/s io/ i ‘I 40-40-40 2 I9 33 I6 8 2 6 23 32 I4 5 2 24 33 4‘I ‘I ‘I 5 I Not clipped I4 23 I0 5 ‘I Average 22 32 I3 6 2 0-0-0 2 ‘I6 20 I I 5 I 6 18 30 7 4 I 24 24 28 ‘l0 2 2 Not clipped 28 32 ‘l0 4 2 Average 2‘l 28 ‘l0 4 2 ‘Clipped to stubble heights indicated on August 20, ‘I962. 25 days after initial shattering. Apparently inflo- rescence production continued even though the plants were not cut and the late developing inflorescences started maturing about 25 days after the first inflo- rescences matured. Since the fall seed crop tends to mature more unevenly than the first crop in the spring, the effect of clipping at various heights in late summer was studied in 1962. Results are presented in Table 2. When fertilizer was applied at the time o-f clipping, yields were improved following clipping. Some seed were mature and started shedding approximately 1 month after clipping. Peak harvested yields were obtained 5 days after initial seed shattering. When fertilizer was not applied at the time of clipping, yields were as good without clipping as with clip-ping. Again, the peak harvested seed yields were 5 days after initial shattering on clipped plots. The yields are- reported as cleaned or pure seed. While the yields are relatively low, the results suggest that frequent harvest is possible since mowing 30 days prior to harvest did not affect yield. Several harvests should result in pure seed yields in excess of 150 pounds per acre. A system of harvest whereby the plants were not cut but rather the mature seed Pounds seed - 40 30g \\<:-—-Percent purity ' 1 20E ’—.- i1 \ _ '- _ ,1 ‘VI \\_____.. I08 I l l l l I l l l I l l I l I l l l l I l l l l I l l l l I l l l l l I l l l l I l l l O I5 20 l0 e I5 2O 25 30 5 l0 June August sePiembef e 5. The influence of time of harvest on seed yield and purity of kleingrass, 1960. Initial sampling (harvest) was on date of 5 removed by agitation at frequent intervals might increase yields still further. This is suggested also by the results in 1960 (Figure 5), in which a second peak in yield was obtained 25 days after initial maturity even though the plants had not been cut. I Introduced Bluestems Pretoria 90 Bluestem Pretoria 90 makes good vegetative growth in mid- summer, but floral initiation and development occur only sporadically prior to October 1. A test was started in 1961 with Pretoria 90 bluestem and in 1962 with Accession 4080 bluestem in which a part or all of the top growth was removed by mowing. The mowing practices were conducted in combination with various fertilization and irrigation practices. On August 20, 1961, Pretoria 90 plants were 36 inches high and showing some indication of booting. Plots mowed to stubble heights of 2 and 6 inches on that date had again reached the boo-t stage on September 20 with inflorescence emergence starting on September 25. Where the plants were clipped to 24 inches, or about 12 inches of the tops removed, sparse heading was noted on October 11. Nonclipped plants were not in full boot until October 14 with some emergence starting on October 17. Inflorescence counts made on all clipped plots on October 17 and 10 days later on unclipped plots are shown in Table 3. Clipping not only resulted in an increase in number of inflorescences but also in earlier maturity of the inflorescences. The later development of inflores- cences with 24-inch clipping is indicated by the reduced yield. While unclipped plots finally headed, the inflorescences failed to mature in 1961; thus, no seed were produced. A similar pattern o-f mowing was followed on the same plots in 1962 and 1963. Inflorescence pro- duction and seed yields followed approximately the same pattern as 1n 1961 (Table Inflorescence counts were made at the same time on all plots in 1962 and 1963. Some inflorescences probably emerged later, but because of their lateness, they did not mature either before harvest or before frost. TABLE s. rue INFLUENCE or CLIPPING PRACTICES ON FALL SEED PRODUCTION o‘; PRETORIA 9o BLUESTEM, COLLEGE STATION, 1961-ea ' Clipping height, inches Year 2 6 24 Not clipped Number of inflorescences per foot of row 1961 24 27 3O 21 1962 24 15 9 5 1963 18 18 14 8 Average 22 20 18 11 Pounds of seed per acre 1961 165 163 108 0 1962 120 100 81 69 1963 86 77 52 37 Average 124 1 13 8O 35 TABLE 4. THE INFLUENCE OF IRRIGATION ON FALL SEED PR1 TION OF PRETORIA 9O BLUESTEM, COLLEGE STATION, 1961- Not Irrigated to Irriga i Year irrigated boot stage to ma Number of inflorescences per foot of 1961 31 23 23 ' 1962 19 1 9f 1963 3 _ A,‘- 19 24g. Pouinds of seed per acre g 1961 159 147 131 1962 9s 1 9| - 1963 0 6O 63 lNot included in 1962 test. These data definitely indicate that surf growth should be removed about mid-August improved fall seed production. ‘ The effect of irrigation on fall seed produc of Pretoria 9O bluestem is shown in Table 4. Q sponse to irrigatio-n depended largely on rai during the fall seed production period. This s was conducted on a shallow, light-textured soil; t: rainfall distribution as well as total rainfall influ response to irrigation. Rainfall for August, S_ ber and October totaled 10.8, 10.7 and 6.7 in respectively, in 1961, 1962 and 1963. Yields wi irrigation were equal or superior to irrigated y in 1961 and 1962, while in 1963 only erratic i res-cence production occurred on dry land, and inflorescences failed. to mature seed. Irrigatiog necessary for consistent seed production in this A fertilizer variable was included in these st but responses were not consistent; thus, yields are; shown. In general, fertilizers are necessary for f plant growth on the light-textured soils, and plant growth is necessary for seed production. least 40 pounds of nitrogen per acre per seed _g should be applied, along with phosphorus and o. sium as indicated by soil tests. I It is possible that close mowing in late suf‘ might reduce vigor and therefore, the following crop. The results presented in Table 5 show if any effect of late summer clipping on seed n" tion the following spring. There was a slight in I in yield and an offsetting decrease in quality 1 closer clipping. Fall irrigation tended to d spring seed production. 31 The spring seed crop averaged 87 pounds» acre and the fall crop 92 pounds, for a total 0 pounds of seed material per acre in 1962; ho I quality was poor in both crops. 1i Seed shattering occurs in the introduced ‘X stems, and stage of maturity for harvest is some difficult to determine. A study was run in 195g v again in 1960 on the spring seed crop to det the rapidity of change in potential seed yield j delay in harvest. Samples were harvested at intervals beginning when the tips of approxii THE RESIDUAL EFFECTS OF PREVIOUS YEAR TREATMENTS t, RING SEED CROP OF PRETORIA 9O BLUESTEM, AGRONO- "1 , 1962‘ Clipping treatment, fall 196T 2-inch 6-inch 24-inch None Aver“ e height height height 9 1 17 1 3O 83 8O 102 f» 11s 10o 64 101 9s “,_ to , s6 6s 7o 67 63 t]. 96 9a . 73 a2 g percent .101 10.0 11.6 12.7 14.9 _, received a 40-40-40 fertilizer application when growth ‘in 1962. Scent of the heads would release seed with a pull between the fingers. It is apparent from ults in Figure 6 that maximum seed yield is ‘ed within 2-4 days of this time. Expected ted yields decrease rather rapidly following the f um period. "on 4080 Blucstem esults with Pretoria 90 indicate that seed pro- _'_0n can be enhanced by certain cultural, manage- and harvest practices. However, potential seed = ction of this excellent forage grass is still i1 Observations and yield trials with Accession f» indicated that it performed similarly to Pretoria 3 forage yield, and seed set averaged 4O percent l_ single-row nursery in the fall of 1961. Thus, ion 4080 bluestem was evaluated for seed pro- '0n since its vigor was approximately equivalent getoria 90. A seed-production block established lows in the spring of 1962 was clipped, watered fertilized starting in August in 1962 and 1963 ‘ ding to the scheme outlined for Pretoria 90. apparent that Accession 4080 responded to "Q ing similarly to Pretoria 90 (Table 6). As in- pity of clipping decreased, both inflorescence pro- ion and seed yield decreased. These results , itely indicate that removal of vegetative growth ,h accumulates in the summer favors fall seed :1 uction. TABLE 6. THE INFLUENCE OF VARIOUS CLIPPING PRACTICES ON FALL SEED PRODUCTION OF 4080 BLUESTEM, AGRONOMY FARM, 1962-63 Clipping height, inches Year 2 6 24 Not clipped Number of inflorescences per foot of row 1962 16 20 12 15 1963 19 23 11 12 Pounds of cleaned seed per acre 1962 67 60 37 49 1963 83 51 54 43 Irrigation had opposite effects in the 2 years on Accession 4080 bluestem (Table 7). Inflorescence production and seed yield were better in 1962 without irrigation, while in 1963, the few inflorescences pro- duced without irrigation failed to mature seed. It is apparent that irrigation is critical for seed pro- duction. It must be available for consistent produc- tion, but if used when not absolutely necessary, irrigation may actually reduce seed yield. Rainfall in 1962 during August, September and October totaled 10.7 inches. In 1963 the total for the same period was 6.7 inches. The plants appeared to show moisture stress at times in 1962 and were irrigated, but apparently before it was actually necessary. Acces- sion 4080 sho-wed a greater reaction to irrigation in 1962 than Pretoria 90. Inflorescence production of Pretoria 90 was retarded, as indicated by counts made in October, but counts at the time of harvest in November were similar for plots with or without irrigation. Furthermore, Accession 4080 inflorescence production was lower on the day of harvest with irrigation. Apparently irrigation favored vegetative development over inflorescence production. Medio Bluestem Medio bluestem sometimes produces an excellent seed crop of satisfactory quality. At other times both yield and quality may be quite poor. Studies similar to those reported for Pretoria 90 and Accession 4080 were conducted with Medio, with the exception of mowing. Summer growth of Medio did not attain sufficient height to permit the mowing treatments. h; 12o - 1959 F‘ A 6 I00 - I’ \-—|9eo 5 \ , 3 8O _ \\ Z Figure 6. Seed yieldqof Pretoria 90 a \ l ' bluestem as influence}? by the length Y, 60 - \_ of time after first maturity until har- 3 Ts‘ J vest, College Station, 1959-60. '5 4Q . 43’ C 3 20 - m ’ o I 1 I 1 1 1 I 1 1 1 1 I 1 1 1 I 1 1 1 1 I 1 1 1 1 I 1_.1 28 I 5 IO I5 2O 25 May June In 1958, seed were harvested from a small plant- ing of Medio in June and December. The June harvest averaged 59 pounds per acre with 17 percent seed set, and the December harvest averaged 15 pounds per acre with only 8 percent seed set. The results in Table 8 show very good seed production in the fall of 1961. Yields were improve-d significantly with irrigation but quality was decreased. Appar- ently, the irrigation delayed maturity, resulting in reduced quality. Total yield in 1962 was 75-100 pounds from two seed crops. Quality of the spring seed cro-p was satisfactory, and the fall crop was not analyzed for quality. These results and experiences indicate that when the environmental conditions are satisfactory, Medio will produce a good seed crop. At other times 25-50 TABLE 7. THE INFLUENCE OF IRRIGATION ON FALL SEED PRODUC- TION OF 4080 BLUESTEM, AGRONOMY FARM, 1961-63 Irrigated to maturity Irrigated to Year Not irrigated boo‘ stage Number of inflorescences per foot of row 1962 22 1 10 1963 2 28 21 Pounds of seed per acre 1962 74 1 33 1963 O 64 47 ‘Not included in test in 1962. TABLE 8. THE INFLUENCE OF IRRIGATION PRACTICE ON MEDIO SEED PRODUCTION, COLLEGE STATION, 1961-62 November, 1961 June, 19621 November, Irrigation 1962 practice Pounds Percent Pounds Percent Pounds seed seed set seed seed set seed No irrigation 188 46 23 24 5O Irrigated to boot stage 199 28 25 27 5O Irrigated to maturity 217 26 26 24 77 ‘Irrigation variable applies only to fall seed crop as irrigation was not used in the spring. TABLE 9. FORAGE PRODUCTION AND SURVIVAL OF INTRODUCED BLUESTEM ACCESSION, TEXAS A&M PLANTATION, 1958-60 Forage yield Smnd (tons/acre) survivd 195a 1959 1960‘ (Pmeml Accession No. or variety name » Specles Pretoria 9O Dicanthium annulatum 5.0 8.8 2.5 50 Medio Dicanthium caricosum 2.2 4.9 1.7 100 Gordo Dicanthium annulatum 2.5 4.9 1.1 40 T-20061 Dicanthium sp. 3.7 6.2 1.3 40 T-20299 Dicanthium sericeus 3.6 6.5 1.4 55 lOnly one harvest was made in 1960 because of poor stand. pounds of seed per acre are obtained two or times per year. Irrigation may be necessary f0 production, depending on the amount of natural; fall, but late fall irrigation may result in re quality thro-ugh delayed maturity. ' Experimental Accessions Numerous experimental accessions of the I duced bluestem complex, mainy of them fro i Oklahoma Agricultural Experiment Station, g been evaluated for various agronomic character, including seed production. The accessions » grown first in single rows and evaluated visua desirable characteristics. The most promising were then included in replicated tests and =5“ for forage and seed produ-ction. Yields of five lines grown in the Brazos bottom from 1958-60 are shown in Table 9. y_ ously, none of the lines are superior to Preto, in total forage production. Medio, possibly o"- of its growth habit, persisted much better tha other types. In 1960, a test was established ll upland farm involving several additional lines v had shown promise in observation rows. The g indicate that most of these were less vigorous, Pretoria 90 (inside front cover table). It noted that most of the less virogous lines were sufi for seed production. These included Gordo, i, 4081 and 3025. All of these had exceptionally‘ seed quality but did not persist in the yield whicg‘ to be abandoned. 5 In 1962 the replicated test was re-establishe weak lines eliminated and additional promising- included. Accession 4080 was approximately to Pretoria 90 in forage production and quality d the 3 years of these tests. Its seed productio =1 quality were better than Pretoria 90 in 196k poorer in 1962 and 1963. Accession 6141, a B ochlo-a species, produced an excellent seed cr! 1961 but was unsatisfactory in 1962 and 1963 forage yield was below the better Dicanthium and forage quality as indicated by leaf perc was very poor. Of the additional lines evalua 1962 and 1963, only Accession T4931 show promise of satisfactory performance. .1 Pretoria 90 has not been produced on a I scale because of its limited seed production. L it is no-t likely that any of the materials eval in these studies would be any more satisf" Among those accessions with satisfactory vigof persistence, only Accessions 4080, 6141 and show promise of being equal to Pretoria 90, an do not exceed Pretoria 90 in seed production.