247-427-9M-L1 8O TEXAS AGRICULTURAL EXPERIMENT STATION B. YOUNGBLOOD, DIRECTOR COLLEGE STATION, BRAZOS COUNTY, TEXAS BULLETIN N0. 359 _ MAY, 1927 DIVISION OF AGRONOMY HYBRID VIGOR IN SORGHUM AGRICULTURAL AND MECHANICAL COLLEGE OF TEXAS T. O. WALTON, President STATION STAFFT ADMINISTRATION: ‘B. YOUNGBLOOD, M. S., Ph. D., Director A. B. CONNER, M. S., Acting Director . E. KARRER, B. S., Acting Vice-Director . M. SCHAEDEL, Secretarig . P. HoLLEMAN, JR., C ief Clerk . K. FRANcKLow, Assistant Chief Clerk CHESTER HIGGS, Executive Assistant C B. N EBLETTE, Technical Assistant VETERINARY SCIENCE: * M. FRANCIS, D. V. M., Chief H. Scm/nm‘, D. V. M., Veterinarian J. D. JoNEs, D. V. M., Veterinarian CHEMISTRY: G. S. FRAPS, Ph. D., Chief; Slate Chemist S. E. AsEURY, M. S., Assistant Chemist WALno H. WALKER, Assistant Chemist VELMA GRAHAM, Assistant Chemist AnAn E. STURGIS. B. S., Assistant Chemist E. C. CARLYLE, B. S., Assistant Chemist R. O. BRooKE, M. S., Assistant Chemist T. L. OGIER, B. S., Assistant Chemist J. G. EVANS, Assistant Chemist HORTICULTURE: W. B. LAINHAM, M. A., Chief H. NEss, M. S., Berry Breeder RANGE ANIMAL HUSBANDRY: J. M. JON_ES,_A. M., Chief; Sheep and Goat Investigations J. L. LUsR, Ph. D_., Animal Husbandman; Breeding Investigations W. H. DAMERON, B. S., Wool Grader ENTOMOLOGY: F. L. TnouAs, Ph. D., Chief; State Entomologist H. J. REINRARD, B. S. Entomologist W. L. OWEN, JR., M. S., Entomolo ist S. E. McGREcoR, JR., Acting Chie Foulbrood Irwector Orro ACKENSEN‘, Foulbrood Inspector GILLIS GRAHAM, Foulbrood Inspector AGRONOMY: E. B. REYNoLns, M. S., Chief A. B. CoNNER, M. S., Agronomist; Grain Sorghum Research R. E. KARPER, B. S., Agronomist; Small Grain Research MANcELsnoRF, Sc. D., Agronomist; Corn and Small Grains _D. T. KILLOUGH, M. S., Agronomist; Cotton Breeding E. C. Cusnma, B. S., Assistant in Crops P. R. JOHNSON, Assistant in Soils hguw PLANT PATHOLOGY AND PHYSIOLOGY: , . J. TAUBENHAUS, Ph. D., Chief FARM AND RANCH ECONOMICS: L. P. GAREARn, M. S., ie *B. YOUNGBLOOD, M. S., P . D., Farm and Ranch Economist G. L. CRAvqFoRn, M. S., Marketing Research Specialist V. L. CORY, M. S., Grazing Research Botanist "*T. L. GASTON, JR., B. S., Assistant, Farm Records and Accounts ***J. N. TATE, B. S., Assistant, Ranch Recordl and Accounts RURAL HOME RESEARCH: JESSIE. WHITACRE, Ph. D., Chief SOIL SURVEY: ***W. T. CARTER, B. S., Chief H. W. HAWKER, Soil Surveyor E. H. TEMPLIN, B. S., Soil Surveyor T. C. RErrcR, B. S., Soil Surveyor BOTANY: H. NEss, M. S., Chief PUBLICATIONS: A. D. JACKSON, Chief SWINE HUSBANDRY: FRED HALE, M. S., Chief DAIRY HUSBANDRY?“ i» —————— ——, Chief POULTRY HUSBANDRY: R. M. SnEnwoon, M. S., Chief ****AGRICULTURAL ENGINEERING: MAIN STATION FARM: G. T. McNEss, Superintendent APICULTURAL RESEARCH LABORATORY: (San Antonio) H. B. PARKs, B. S., Apiculturist in Charge A. H. ALEx, B. S., Queen Breeder FEED CONTROL SERVICE: D. FULLER, M. S., Chief . D. PEARcE, Secretary . H. ROGERS, Feed Inspector . H. W000, Feed Inspector . L. KIRKLAND, B. S., Feed Inspector W. D. NORTHCUTT, JR., B. S., Feed Inspector wéumg SUBSTATIONS No. l, Beeville, Bee County: R. A. HALL, B. S., Superintendent No. 2, Troup, Smith County: W. S. Horcnmss, Superintendent No. 3, Angleton, Brazoria County: R. H. STANSEL, M. S., Superintendent No. 4. Beaumont, Jefierson County: R H. WYcRE, B. S., Superintendent No. 5 , Temple, Bell County: H. E. REA, B. S., Superintendent N736, Denton, Denton County: P. B. DUNKLE, B. S., Superintendent No. 7, Spur, Dickens County: _ R. E. DICKSON, B. S., Superintendent No. 8, Lubbock, Lubbock County: D. L. JoNEs, Superintendent FRANK GAmEs, Irrigationist and Forest Nurseryman No. 9, Balmorhea, Reeves County: J. J. BAYLEs, B. S., Superintendent No. 10, Feeding and Breeding Station, near College Station, Brazos County: SRERwooD, M. S., Animal Husband- man in Charge of Farm L. J. McCALL, Farm Superintendent No. ll, Nacogdoches, Nacogdoches County: H. F. MoRms, M. S., Superintendent ***No. 12, Chillicothe, Hardeman County: J. R. QUINBY, B. S., Superintendent ***J0sEPn C. STEPHENS, M. A., Junior Agronomist No. 14, Sonora, Sutton-Edwards Counties: E. W. THOMAS, B. S., Superintendent ‘N. L. BLACK, D. V. M., Veterinarian V. L. CoRY, M. S., Grazing Research Botanisl "*0. G. BABcocK, B. S., Collaborating Entomologist - O. L. CARPENTER, Shepherd No. 15, Weslaco, Hidalgo County: W. H. FRIEND, B. S., Superintendent M. McPHAiL, B. S., Entomologist No. 16, Iowa Park, Wichita County: E. J. WILSON, B. S., Superintendent Teachers in the School of Agriculture Carrying Cooperative Projects on the Station: . W. ADRIANcE, M. S., Associate Professor of Horticulture G a. W. BILSING, Ph. D., Professor of Entomology . P. GROUT, M. S., Professor of Dairy Husbandry V. P. LEE, Ph. D., Professor of Marketing and Finance D. ScoATEs, A. E., Professor of A ricultural Engineering _ _ H. P. SMITH, B. S., Associate Pro essor of Agricultural Engineering 'I'As of l\lay 1, 1927. *On Leave. _ _ _ "Dean, School of Vetennary Med1c1ne. '**In cooperation with U. S. Department of Agriculture. "***In cooperation with the School of Agriculture. SYNOPSIS This Bulletin discusses the results attending the crossing of different strains of sorghum of the same variety with one another and of different varieties with one another. Studies of this character are of particular interest to plant breeders in connection with the development of new and superior strains and also to the grower in the practice of growing planting ‘seed. The data presented show that crosses between difierent varieties of sorghum exhibit marked hybrid vigor as meas- ured by the height of plant. These crosses also exhibit un- usual hybrid vigor in the size of leaves, chlorophyll develop- ment, and production of grain, and are characterized by a marked delay in the time of maturity. On the other hand, crosses between strains of the same variety either do not exhibit vigor at all or display it only in the height of the plant. The practical application of the data presented lies in the fact that hybrid vigor may be carried over into succeeding generations and may, therefore, be of practical importance to plant breeders in securing improved strains by obtaining recombinations which retain desirable growth factors. It will also be of value in the study of the relationship of the differ- ent varieties of sorghum. It further alfords a basis for the farmer to arrange his seed fields in such a way that in the succeeding crop rogueing can be practiced with the certainty of eliminating first-year hybrids. CONTENTS PAGE Materials and Methods Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Height of Parents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Crosses between Different Varieties . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Crosses Between Strains of the Same Variety . . . . . . . . . . . . . . . . . . . . 12 Hybrid Vigor in the Improvement of Sorghum . . . . . . . . . . . . . . . . . . 18 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 I ity of neighboring families by cross-pollination. BULLETIN No. 359 MAY, 1927 HYBRID VIGOR IN SORGHUM A. B. CONNER AND R. E. KARPER Texas Agricultural Experiment Station The study of hybrid vigor or heterosis-is a matter of common inter- est to the student of heredity and to the breeder. Maize furnishes a most interesting example of hybrid vigor, and the knowledge obtained affords a basis for the practical utilization of it by breeders in the production of seed corn. In grain sorghum, no such practical advantage seems possible on account of the fact that it is difficult t0 obtain first-year hybrid seed in quantity sufiicient for field use. In studying hybrid vigor in sorghums, therefore, the plant breeder is mainly interested in (l) the increased vigor which can be retained in later generations by the recombination of favorable growth factors, (2) the relationship of different varieties as indicated by the degree of heterosis which follows their crossing, and (3) establishing a basis for eliminating natural hybrids from fields of pure-bred varieties grown for seed. A study of hybrid vigor in difierent crosses opens up to the breeder a better working knowledge of the complexity of the hereditary mech- anism with which he is working. In maize, a crop which is largely cross-fertilized, one finds that within a variety the various families rep- resented have crossed and inter-crossed to such extent that the genetic constitution is very comp-lex. On the other hand, in_sorghnm,g.a_crop which is largely close-f9Ltilized,-one finds that within a variety a num- ber of families or strains are represented in practically pure form, each retaining its identity without contributing a great deal to the complex- The smallness of the amount of natural cross-fertilization is perhaps responsible in ‘a large measure for the ease with which new strains can be isolated by selec- tion, and, on the other hand, it probably plays no small part in con- tributing toward making it relatively easy to recover from hybridized ties or strains of grain sorghum with one another. here relate exclusively to height measurements, and since height is families strains that are homozygous in a sufficient number of charac- ters to be acceptable to the layman as a distinct variety. A knowledge of hybrid vigor, therefore, in sorghum should contribute toward a bet- ter understanding of the genetic complex of the sorghum plant and promote the use of more effective breeding methods. It is the purpose of this Bulletin to present data obtained at the Lubbock Substation from crossing different varieties of grain sorghum with one another, and, as wrell, data obtained from crossing sub-varie- The data presented 6 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION known to be one of the reliable measures of hybrid vigor, these height measurements furnish a means of estimating hybrid vigor in crosses. Materials The materials used in this study include the three strains of milo known as Extra Dwarf, Dwarf, and Standard; and three strains of feterita known as Extra Dwarf, Dwarf, and Standard. It is note- worthy that in both milo and feterita parallel forms have been devel- oped in so far as stature is concerned. The mean height for Extra Dwarf milos and feteritas ranged from 56 to '72 centimeters. The Dwarf forms of milo and feterita used ranged from 94 to 110 centi- meters in height, and the Standard forms of milo and feterita used ranged from 125 to 147 centimeters in height. The Extra Dwarf, as well as the Dwarf form of feterita, has been developed by the Texas Station. All of the materials used were uniformly true to type in so far as height is concerned, as is shown by the parental heights in the tables. None of the varieties used, however, were directly from bagged seed. In other words, the parental material as used in this work was from open fertilized seed representative of the varieties as kept pure by the ordinary methods. Methods When the crosses used in this study were made, the practice was fol- lowed of bagging the two parent heads together in the same bag and shaking the bag occasionally from time to time to more thoroughly dis- tribute the pollen. This resulted in cross-fertilization of a small per- centage of the seeds in the bagged heads. All of the seed of the bagged heads were planted in the field and the hybrid plants were identified by the increased vigor which they displayed, or the increased height over the selfed’ seed. The dwarf strain was used as the mother parent, as previous experiments had shown that F, plants could be recovered with certainty by reason of heretosis or by the display of simple dom- inance of the tall parent. These first-generation hybrid plants were subsequently planted in head-rows for further verification and meas- urement. During the season of 1922, the first-year hybrid plants were meas- ured and their heights were compared directly with the mean heights of the parent varieties grown the same year. In 1923, the second gen- eration rows were compared directly to the height measurements ob- tained from the parent varieties grown in that year, so that a reasonably accurate measure of both the first and second generation material was obtained as compared to the height of the parent varieties grown in the same season. Height of Parent Varieties A description of the parent varieties, in so far as height is concerned, is of basic value in the analysis of data to be subsequently presented. HYBRID VIGOR IN SORGHUM 7 Moreover, it is desirable to have a measure of parent varieties grown under the seasonal conditions obtaining in both 1922 and 1923, the two years during which the first- and second-generations were given. Accordingly, the means, standard deviation, and coeflicient of variation are presented for height of plant of the varieties used. No measure- ments on the Extra Dwarf varieties were obtained in 1922; however, their measurement for a two-year period is not so necessary, inasmuch as we are concerned primarily with the taller parent. Table 1.—Height Measurements in Cm. of Parent Material in Seasons of 1922 and 1923. No. of Coefficient T. S. Name of Variety Measure- Mean Standard of No. ments Deviation Variability (1922) 6312 Extra dwarf feterita . . . . . . . . . . . . . . . . . . . . 65* . . . . . . . . . . . . . . . . . . . . . . . . 4222 Extra dwarf yellow milo . . . . . . . . . . . . . . . . . 72* . . . . . . . . . . . . . . . . . . . . . . . . 5985 Dwarf feterita . . . . . . . . .. .. . .. 40 110.00 =1=1 .09 10.23='= 77 9. =1= 70 670 Dwarf yellow milo . . . . . . . . . . . . 40 104.25 =1=1.64 15 .36 i1 16 14.73 =1=1 13 1652 Standard feterita . . . . . . . . . . . . . 40 147.75 =1=1 .38 12 .94=1= .98 8 .76 =1= .66 3301 Standard milo . . . . . . . . . . . . . . .. 30 144.17 =1=2 .67 21 .72 11.89 15 .06 =1=1 .34 (1923) 6312 Extra dwarf feterita . . . . . . . . . . 117 56.955: .23 376i .17 6.60 =1: .29 4222 Extra dwarf yellow milo. . . . . . . 105 65 .69 =1= .55 8.05 =1= .39 12 .25 =1= .56 5985 Dwarf feterita . . . . . . . . . . . . . . . 104 96.46 =1= .51 7 66 =1= .36 7.94=|= .37 670 Dwarf yellow milo . . . . . . . . . . . . 84 94 .79 =1= .57 7 68 =1= .40 8 .10 =1= .42 1652 Standard feterita . . . . . . . . . . . . . 17 125 .50 =1=2 .02 12.37 =1=1 .43 9 .85 =1=1 .14 3301 Standard yellow milo . . . . . . . . . 21 126 .93 =1=1 .99 13 .55 =1=1 .41 10 .68 i1 . 12 ‘tlnterpolated. _ All the varieties, under seasonal conditions in 1922, were taller than they were in 1923. (Fig. 1.) This fact is one to be considered later in connection with the heights of the first-generation crdsses. Crosses Between Different Varieties Three crosses were made between different varieties as follows: I. Extra Dwarf feterita and Extra Dwarf Yellow milo. II. Extra Dwarf Yellow milo and Dwarf feterita. III. Extra Dwarf feterita and Dwarf Yellow milo. Cross I is between two Extra Dwarf forms of different varieties, the mean heights of the two varieties differing by only 9 centimeters. The table following shows the frequency distribution and statistical con- stants for the two parents, the first generation, and the second gener- ation. The mean height of the first-generation plants in 1922 is much greater than the height of the two parents as measured in 1923. The feterita x milo cross shows an increase in height of 130 per cent over 8 BULLETIN NO. 2.2.3; FA Ea fih 2i s23 3:53 Q0 cowimnrcou wan “N2 vcm mmafi E ECBmE 135:3 Mo“ Hflfim Ho gamma: 53E 23 M52225‘; $74 359, TEXAS AGRICULTURAL EXPERIMENT STATION 3Q Q g $9 Q l. fixxkNmwk Q .\§\ . ER, Q§§~§Q .353 .SR\\.§Q khQg ufiiax». X X X X X X xwfi§xw .o§§\§- Q§§§QRYE§~R9Q $§~k§§ Fw¢§Q aw 0 m w. o9 QN\ 0% HYBRID VIGOR IN SORGHUM 9 Fig. 2.—Dwarf milo (670), left; oxtra dwarf feterita (6312), right; _and F1 hybrid, 'center, showing marked hybrld vigor accompanying Wide crosses 1n sorghum. 1O BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION sqfiwfi .=.1$m.mm 3131”: ma H m a 2.. a g Nu 2 N. Q ma: wmfiflsd $12.2 mimfigwfi w _ H .. é 3: a: H33 8. i: 5 Haw; i . . .. 2 a. 2 m m ma: QQHQZ. zflmsam mwiim E Ma m E: miflmvfi figflmfi 3.1322 a“ .. . . . .. m $2 .2 . . .. .. . m $2 2N 18 m2 T: m: m2 a: m3 T2 5 m: n: 3 _ 3 _ 2 i 2 _ 2 nofingvQ HMNQE .02 QUE. HN®> §O@¢gw@wQQ .@O QEQTwEQOO .@.-N.Tifiw@ éfiwOrf LNMQCQU Oz .@ J? HQN§% %O Mpw MkQwQ-HMEQHQO HE mhflnwfiwnv .553 0E2 wbfififi wan Agmwv wthivh YmBQ mbxm couzfiom wmoHU E ESQ we EwEE To ioflnnwflmmfi hucosvofimlq“ 05mm. A: mmomu S. “v28 my. “Sim 3. Higw: ME m m w 3 w“ 3 mm Ms Ma 3 a. S Q mm M3 v Q ma: £3 x 2a. ivummm...“ moafimmd wwguuccafi w _ Q ma? nwamxmwmv 5. n35“ 3. H3.» S. U315 i: . mm 5. mm fi .. m £2 . . . . . . zmmam 3. H32 3 “B; 2. fiafi m2 m mm 3 S m £2 . . . . . . Qmfi 8N T3 T: m5 T3 T3 TE m2 m2 m: 2: 2w i 3 W! 3 l S zrmnfihfiw nosfikwfl s52 o Z cos aavw aofiacwmmwfi To "Euéfiwoo wéwafim Eon. épmzmw d2 .m H £55 we 39mm .6“ BfiwazucwO E mkfinwO mmfio Amwmmv a253,.» $55G was Ammmwv 0E2 wuflk/Q whim cwvzfiom wwohU E “Ewqm Mo wsmmum we cofisfibwmfl mucwzvohblh wEmH. .2 mwomu Qjflzlm 3. nmfiYmm alflmwofi 2: H m 3 on 3 E S 2 mm E 2 Q $3 N5 w Nwfi mm. fix? 3 "$2M" E. HRamS a: m w h mm S E S my E. 3 my Q Q i $3 mg. x N5. Nimulwfi fiwmfiii B..L.HQmm: m i $2 fimiwfimw S. Hal wwluwfi @w.13§m_ h . m _ M . . PM $2 Nwfi x 2% 3. H32 2. H2; 3 HQZQ m2 i L“ m ma: . . . . . . zga 3. Hal S. Hfiam 3 H215 >3 . .. m Mai . zfimw 8m T3 T: n2 T3 T3 TE n2 m2 TS T: 2.. i 3 fimznwia> mo$fi>~Q n32 dZ nos .5»? nozknwwwwfi we fiiofiwoU wiwnfiw 13cm. LfiQEU dZ .m Q. aim Mo 22mm 8w MHSBEQQQO E. Efinvv $20 $95G whim $56 mfihzoh “fig/Q wfixm nova/Em mwOhQ GM HGN~Q MO Qammom m0 coSnnmSmmQ ~WOGOUUQMHMII.N MJQNH. A wwomo HYBRID VIGOR IN SORGHUM 11 Extra Dwarf milo, the taller parent. The reciprocal cross shows an increase of '74 per cent over the taller parent. The difference in height between the reciprocal crosses is greater than three times the probable error, but the sample is so small, seven plants in one case and five in the other, that little significance can be attached to this observation. It will be noted that the above comparisons are based on measure- ments of the F, plants in 1922 and parental types in 1923. Unfortu- nately no measurements of the parental types are available for 1922. Assuming that this relation between the extra dwarf and the standard types was approximately the same in 1922 as in 1928, we can inter- polate the height of Extra Dwarf feterita for that year at 64.95 cm. and the height of Extra Dwarf milo at 72.25 cm. A comparison of the calculated height of these two parental types in 1922 with the F, hybrids the same year indicates that the hybrids were approximately double (Fig. 2) the height of the taller parent. The second-generation progeny shows a mean height of 114, as com- pared with 65 centimeters for the taller parent, the measurements be- ing directly comparable. A considerable amount of hybrid vigor has been carried over into the second generation and shows an increase practically equivalent to that shown in the first generation. As would be expected in the second generation, the variability is strikingly large. Cross II is between Extra Dwarf Yellow milo and Dwarf feterita, two varieties that differ in height by about 30 centimeters. Table 2 shows the frequency distributions for the two parents, as well as for the first and second generation. In comparing the mean heights of the parents with the mean height of the first-generation progeny, one sees that a marked increase in height takes place in the first generation of this cross. A direct com- parison for height of the first generation with the taller parent shows a difference of 2'7 centimeters in height in favor of the first-generation progeny, or an increase of 2'7 per cent. The distribution indicates that, while marked hybrid vigor accompanies this cross, the expression is not so pronounced as in cross I, when the two Extra Dwarf varieties were combined. Vigor extending over into the second generation seems to be fully as prominent as it was in the first generation. CYOSS III is another inter-varietal combination between Extra Dwarf feterita and Dwarf Yellow milo, varieties that. differ in stature by 38 centimeters. The results of this cross, as judged by the heights, are shown in Table 4. ‘ These two different varieties, when crossed, show a first generation with a mean height of 146 centimeters; that is, they are approximately one-half taller than the parents. An increase of 41 per cent over the tall parent is shown in the first generation and an increase of 25 per cent in the second generation. While this cross is attended by marked hybrid vigor, it seems to be carried over into the second generation in a lessened degree. If the dwarf and extra dwarf forms in both varieties differ from 12 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION each other by only a few factors for height and are identical in the re- mainder of their genetic factors, it might be expected that the F, hybrids which result from crossing sub-varieties of milo with, those of feterita would attain approximately the same degree of vigor, regard- less of the sub-varieties used as parents. In other words, it would seem reasonable to suppose that the F, hybrid betwen the two extra dwarf forms would be identical to the hybrid between a dwarf x extra dwarf, a dwarf x standard or even standard x standard. Though the data show that there is some fluctuation in the height of the F‘, plants from the three inter-varietal crosses, the variation is no greater than might well be attributed to the effects of environment and the small size of the sample. . In the F2 generation, where larger populations were available, the average height is practically the same for every cross, being 113.77 and 1120.84 for cross I and 119.11 and 118.49 for crosses II and III, re- spectively. The coefficients of variability are likewise very nearly the same, 28.58 and 27.01 for cross No. I and 25.35 and 29.84 for crosses II and III, respectively. These facts would seem to indicate that the three crosses represent approximately the same combinations of genetic factors, and that the height differences between sub-varieties of the same variety are not effective when inter-varietal crosses are made. Although detailed measurements are not available, it may be men- tioned that the hybrid plants resulting from these three crosses ex- hibited heterosis in many characteristics other than height. The stalks were larger and more leafy; the color a deeper green; the heads and plants were heavier, and, unlike hybrid corn plants, considerably later in maturity than either parent. Crosses Between Strains of the Same Variety Three crosses were made between strains within the same variety. These were as follows: .IV. Extra Dwarf Yellow milo and Standard Yellow milo. V. Extra Dwarf Yellow milo and Dwarf Yellow milo. VI. Extra Dwarf feterita and Standard feterita. These different strains within varieties have been selected because of inherent differences in height and are distinct in respect to this charac- teristic. (See Fig. With the exception of height of plant, the _ strains within each variety are a counterpart of each other, being iden- tical in other respects. Their genetic constitution is undoubtedly very similar with theexception of the factors for height. . CPOSS IV is a cross of the Extra Dwarf Yellow milo and the Standard, or tall Yellow milo. These two strains differ in height by 61 centi- meters; or, in other words, one is approximately twice as tall as the other, and they afford good material for the study of hybrid vigor. Table 5 shows the frequency distribution and means of the parents, HYBRID VIGOR IN SORGHUM 13 Fig. 3.—Extra dwarf milo (4222), left; standard milo (3301), right; with F1 hybrid in centen N0 hybrld vigor accompanies this cross. Height difTerenee between F1 and tall parent shown here is due to sampling and not slgnlficant. 14 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION duwwixiwwcw» ww "lira. S H3 om N». H03? www w § om S. m. ow Q a mm o "m £2 2w x ma? wwaumwww 33MB N. wwauflowwfi v H m M . . 5 $3 o3 x $4. 3 H8 w 3. H8 A s. Hmfiiw aw . ow mu. 2 m. m ma: . . . . . . 5s ww. H32 ow. H8 w ww Howww m2 .. . w ww $ R H m ma: . . . . . . INN? mfifinmv: @<_H@m.m_ xlfimwx: 3 . .. m ma: . . . . . . 5% . . . . . . . . . . . . . . . . . . . . . . . . . . .. E» m $3 Iitimmfi. wow 2: m: w: m: m2 W1 @121 a a a mW ww @~ ww ww 3 xfisnnixx» floflwwgmg amoz o Z E5 bwww cofiwswowvfl we wnvoomwwoO wbwwnxow E89 éhiww dz .m F p.85 we wnwfim 8w Efiwflwauo E mxwwnwo x35 .65 2:2 EQEQ E5 6&3 Q52 15.5 wbxm Isafiwm $80 E 22m 3 awwsm u. cosiosmma xaoaxnvfimlxo 293w .> mwomu dfifioaxvwuw. S. H58 an H35 ow. “v3.53 wow w D m. x: mo w: mo o» a" ww I m Q $2 3mm x mm? maxi: mwxnwzxw omaflmvxxx h .. .. w m .. é g: 3?. x 3% 213w 2 alumni: xffimwwwfi G M w w N. w H .. m moi . . . . . . 35mm ww. Hfifi ow. Hwow ww nmowww 2: .. .. w wm fi 5 x m £2 . . V . . . zfifi wwAHwowfi gaiva s N333 ow .. m $2 . . . . 4 . 15mm . . . . *2. m $3 4l>ulnrimmmv wow m2 a: T: T: n: TE wfi ma m: m: wo o ww w“ j ww i ww 2 mwzmnfixfw nomwaivQ =82 dZ new xwww nofiwnwwmwfi we EQMQEQQO wemwnwwm 139B -E2$U d2 .m Q. add_% m0 5w fiowwflmwnuO GM mkwwnwO $20 .238 .52 E6923 Es 8mg 2:2 imam wsxm =$atm mwfiu 5 23E u. wwwsm m. Eiimseo xémévfiml...“ 03$. <5 wwomu 15 HYBRID VIGOR IN SORGHUM fiQHQHOQHQQHHMMH 5.13m 3 wNHHE Hm HwHmuwHYwoH 3H H .. H mH Hm Hm wH h w H H. mm NH PH E: S2 x N5. . . . . . . . . . . . . . . . . . . . . . . . . . . .. omdmH H . H . SH uni $3 x ~35 Hfifinmw a QYHHS NH 8 NHEEQ “H . H H m m. w H m $2 . . . . . . rfifi 3. “Is H“ “H. HE m mm "v3.3. SH . .. 1 wH Ma H nH 3: . . . . . . :28 3. HE. w w? H32 H; HHEaSH 3 . m $2 . . . . ....mm@H . . . . . . . . . . . . . . . . . . . . . . . . . . .. ¢$ m wgH ........wHmw 8N 13H T2 fiH 5H 3H 3H T3 3H 75H 3H i 8 3 i E. m. i 2 3 EHHHHHEHQH» nosmtéfl 5w v2 .0 Z mos hnww noHHmswHmonH .6 HHBHQEHEO whmwzwaw HEPH. -3260 dZ .m .,H. 3.2m .3 0.3mm 8H Hzfiwgaawo E. mhwvnwo H620 .89.: wHHHvHoh @5235 Ea firm: “Hrfiom $53G mbxm avofiwm mmoHU E HnmHnH Ho HHHmHoHH “Ho =oa=£b2Q .8=...=v2ml.> 2Q? .H> mmomu 16 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION and of the first and second generations, for height. No heterosis was exhibited in either the first 0r second generation of this cross. The F, generation, based on a small sample, is partially intermediate be- tween the two parent heights in the same year, being 24 cm. shorter than the tall parent. Cross V is between Extra Dwarf Yellow milo and Dwarf Yellow milo, two varieties apparently identical in every respect with the exception of height of plant. These parents differed in height by 20 centimeters, with the individuals in the tall classes of the dwarf parent overlapping with the shorter classes of the tall parent. Table 6 shows the distri- bution for the parents and the first- and second-generation progeny of this cross, together with the means, standard deviation, and cofficient of variability. These two parents, although strains of the same variety, show rather marked hybrid vigor as measured by the height of plant, especially in the first generation of the cross. Although only four individuals were measured in the first generation, the increased height exceeded the tall parent by 20 per cent. The difference in height as exhibited between the tall parent and the F, generation grown in 1922 seems to be sig- nificant, being seven times as large as the probable error. The vigor in the second generation shows a marked decrease over that shown in the first generation and is an increase of 6 per cent over the tall parent. The difference in this generation, with a population of 366, is also sig- nificant, being six times as large as the probable error. In the cross between Extra Dwarf milo and Dwarf milo, height is apparently due to several factors, the increased height shown in the F1 being an expression of two or more complementary height factors, These two strains of milo seem to be identical in their genetic make-up with the exception of the factors for height. N0 other evidences of hybrid vigor were apparent. In Cross VI, the parents involved are Extra Dwarf feterita and Standard feterita, being two strains of the same variety differing widely in height of plant. The tall parent exceeds the shorter by 69 centi- meters, or is more than twice the height of the short parent. This cross between Extra Dwarf feterita and Standard feterita, two extremes in y the feterita class (Fig. 4), affords a comparison with a similar cross, No. IV, between Extra Dwarf milo and Standard milo. The measure- ments on this cross are shown in Table 7. . The first generation shows an increase of 5 per cent over the tall parent, while the second generation shows an actual decrease of 13 per cent from the average height of the tall parent. Extra Dwarf feterita and Standard feterita differ genetically probably only in height. This difference has been found in unpublished data to be due to a single factor. The E1 generation in this cross, therefore, merely exhibits sim- ple dominance of the tall parent, the slight apparent increase in height over the tall parent probably being due to environmental influences afiecting the single F, individual making up the sample. The decrease HYBRID VIGOR IN SORGHUM 17 Pig. 4.—Extra dwarf feterita (6312), left; standard feterita (1652), right, with F1 hybrid in center. ‘ N0 hybrid vigor accompanles this cross. 18 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION in height shown in the second generation is due to the recessive dwarfs in this population and it will be remembered that, on the average, the plants were not as tall in 1923 as in 1922 because of adverse environ-p; mental conditions. Previous crosses between these two strains of _, feterita, when larger populations of both the first and second genera- . tions were recorded for study on segregation of height ofplant, have not shown hybrid vigor to be perceptible in either of the first or second Y generation of this cross. f These three intra-varietal crosses differ from the inter-varietal crosses» previously described in displaying heterosis, if displayed at all, only, with respect to height. No differences between the parents and the y: hybrids were noted with regard to size of leaves, chlorophyll develop-ll ment, production of grain, or time of maturity. Whatever degree of f, heterosis has been exhibited in these crosses has apparently been due to; the interaction of a few genetic factors afiecting only the development; of stature, while in the inter-varietal crosses, interaction of genetic fac-Q tors afiecting many developmental processes was apparently at play. Hybrid Vigor in the Improvement of Grain Sorghum It would seem from the preceding data and discussion that marked? hybrid vigor attends crosses between widely different groups or betweenfl different varieties of grain sorghum, whereas, with crosses between’; 1 strains of the same variety, hybrid vigor is not expressed, or to a1 smaller extent, depending upon the genetic factor differences involved.» '3 Environmental factors were by no means under controlled conditions, a’ I. I. and, furthermore, this work extended over two seasons with rather, widely different effects upon the character used in this instance as a- _» There is, therefore, consid-‘ f erable fluctuation shown in the height of plant between the first threei 5 crosses when strai-ns of feterita, differing widely in height, were crossed This fiuctuy? ation is especially apparent in the first-generation hybrids, while a? measure of hybrid vigor, namely, height. with strains of milo, showing wide differences in height. similar degree of variation is not shown in the second generation. It}, is possible that, under controlled environmental conditions, the exhibiq tion of hybrid vigor in the first generation of these first three crosses} would be more consistent. .3 In the growing of increase fields of pure seed, the breeder and grower; alike are confronted with the problem of natural cross-pollination as} the pollen from sorghum is carried by the wind, and absolute field isoa; lation is difficult to obtain. In the event that nearby fields of otheri; strains exist, the problem of cross-pollination in the seed fields becomes; increasingly difiicult. I by the arrangement of his seed fields so as to have the two strains or‘; varieties flower at different times. Hybrid vigor in sorghum can be‘! used, however, to great advantage by the farmer who plants improved .1 l The breeder or grower of a seed field can Vi“ ’ nothing to avoid cross-pollination other than to use every means possi-i 5 ble to isolate such fields either by the separation from other strains I W1 " HYBRID VIGOR IN SORGHUM 19 seed as ‘a basis for the elimination of any first-year hybrids that occur in the field provided the breeder 0r grower of the seed from Which his field was planted had taken the precautions necessary to have these hybrid plants express themselves in the form of hybrid vigor, thus al- lowing them to be rogued out of the field and entirely eliminated be- fore they have caused any further contamination. This is possible in an effective manner, however, only where the grower has used precau- tion in planting adjacent only those varieties which when crossed are known to express hybrid vigor in the first generation." This simply means in practice that the breeder or grower of increase fields of sorghum should not plant strains of the same variety of sorghum ad- jacent, and that different varieties may be planted adjacent in seed fields with the certainty -that any cross-pollination that occurs can be removed in the succeeding crop. The broader use of pure seed of strains free from contamination, therefore, is dependent upon the co- operation between the breeder, the seed grower, and the farmer who plants the seed. The result of hybrid vigor in first-generation crosses between differ- ent varieties of sorghum is a marked increase in height of plant, size of leaves, and a corresponding increase in the size of the plant, includ- ing its yield of grain. (Fig. 5.) This increased yield due to hybrid vigor cannot, however, be put to economic advantage in sorghum, as is the case in corn, because of the difficulty in obtaining sufficient amounts of first-generation seed for field planting. The non-uniformity obtained in bulk second-generation seed is so great as to make the use of even second-generation seed, which retains some hybrid vigor, not acceptable for field planting by the farmer except possibly where the crop is used for silage. Hybrid vigor is shown to be persistent in a considerable degree in the second generation, and has been observed to exist in grain sorghum in succeeding generations. The principal value attached to this phe- nomenon seems to be the unusual opportunity that exists for the breeder to select out of crossed materials recombinations, in later generations, which have inherited genetic factors contributing to hardiness, produc- tivity, uniformity, earliness, and other desirable characters. Such re- combinations recovered from Wide crosses by the writers and now in the fifth and sixth generations show marked retention of hybrid vigor, even though it is not nearly so pronounced as in the first and second generations. More than fifty such recombinations have been grown and tested at the Lubbock Station, during the past five years, and they show a surprising uniformity, offering to the breeder unusual opportunity for obtaining new and desirable strains with stability of characters in such a degree as to make these recombinations entirely acceptable to farmers as varieties. There are many varieties of grain sorghum in existence today which are known to be of chance hybrid origin, and probably many others not recognized now as of hybrid origin have come into existence by this means. 2O BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION Q $»J L» , ....,...$ vqani i. s... v :12» 4 .7. L4,... in increase Fig. 5.——Crosses between widely related varieties 0f grain sorghum show marked he first generation due to hybrid vigor. standard blackhul kafir right, and F1 below. wei ht of plant and yield of grain 1n t Head of dwarf vellow milo left . 7 v HYBRID VIGOR IN SORGHUM 21 Hegari has every characteristic of a hybrid between kafir and feterita, and although its origin is unknown, it is undoubtedly of hybrid origin. Sorghums cross-pollinate naturally to the extent of about six per cent under favorable conditions so that there is ample opportunity for the origin of natural hybrids. Many varieties which have appeared sud- denly, such as Darso, Maizo, Shrock, and others, are from chance crosses. From these crosses have been recovered recombinations good enough to be considered varieties. Most of these have retained some degree of hybrid vigor. They. are often good producers, but being of unselected parentages and in many cases crosses with sweet sorghums, they frequently inherit undesirable qualities, such as a high tannin content in the seed coat, as well as desirable. Premo and Chiltex, origi- nating at the Chillicothe Station, are varieties which have been devel- oped from artificial cross-pollination and they furnish examples of re- combinations which can be removed from crosses of sorghum. There is a great diversity in types and variation of type from certain crosses, and especially those between different varieties. This great diversity of material possessing hybrid vigor which is carried over into succeed- ing generations offers unusual opportunity of originating and estab- lishing new varieties in sorghum. In crosses made between two inbred lines of kafir which had under- gone eight generations of inbreeding no apparent evidence of heterosis was discerned, and, moreover, ten generations of inbreeding within a number of lines of kafir showed no perceptible reduction in vigor. Since crosses herein reported between closely related varieties, such as Dwarf milo and Standard milo, also do not display increased vigor, it would seem that any advantage in breeding or improvement of the crop which is to be gained by this phenomenon will have to. come from wider crosses or crosses between groups more distantly related. If the degree of expression of hybrid vigor may be taken as a meas- ure of the relationship of the different varieties or groups of grain sorghums, this factor maylend itself to the working out of a better and more adequate key of classification than we have at the present time. For instance, feterita, having somewhat larger and more flat- tened seeds than kafir, is placed in the durra group along with milo. But hybrids between feterita and milo display unusual vigor, indicat- ing that the two varieties differ in a great many genetic factors while hybrids between feterita and kafir are accompanied by hybrid vigor in a very much less marked degree. On the other hand, crosses between kafir and milo also show unusual vigor, frequently accompanied, however, by a decided lack in germination of the seed and by partial sterility, in- dicating, perhaps, an even more remote relationship. Some selections propagated into the third and fourth generations showed a range of sterility from five per cent to ninety-five per cent. 22 BULLETIN NO. 359, TEXAS AGRICULTURAL EXPERIMENT STATION Fig. 6.—Crosses between closely related varieties or strains of the same variety of grain sorghum do not result in increased s1ze of the plant or head. Blackhul kafir left, red kafir right, with four F1 heads of this cross below showing dominance of the red kafir type of head but lack of increase 1n size over the parent. HYBRID VIGOR IN SORGHUM 23 SUMMARY AND CONCLUSIONS 1. A series of three varieties or strains each of milo and feterita having a corresponding common distinctive height and known as Extra Dwarf, Dwarf, and Standard afforded excellent material for measuring heterosis by the height of plant in crosses. 2. Marked hybrid vigor accompanied inter-varietal crosses between milo and feterita in both the first and second generation. In the three crosses between different varieties, the first generation showed an aver- age increase of sixty-six per cent in height of plant over the tall parent. The corresponding second generations gave an increase of forty per cent over the tall parents. ' 3. Crosses between strains of the same variety where the parents differed apparently only in height of plant showed no hybrid vigor where the factors for height were due to a single factor and of the same allelomorphic pair. In a cross between Extra Dwarf and Dwarf milo complementary factors for height give an increase over the tall parent in both the first- and second-generation progeny. 4. An understanding of the expression of hybrid vigor in crosses between different sorghums is of value to» the grower in keeping his seed pure and can be used to advantage in rogueing fields of pure strains. 5. Recombinations of desirable characters in individuals retaining hybrid vigor in later generations offer opportunity for breeding and improvement through hybridization and selection. 6. The degree of heterosis displayed when different varieties are crossed may be useful in determining the relationships of different types and in developing more adequate systems of classification.