329- 1023-11,000-L180 TEXAS AGRICULTURAL EXPERIMENT smnom AGRICULTURAL AND MECHANICAL COLLEGE OF TEXAS I/i W. B. BIZZELL, President gnuunrm mo. s11 SEPTEMBER, 192s DIVISION OF ANIMAL INDUSTRY T he Influence of Individuality, Age, and Season Upon the Weights of Fleeces Produced by Range Sheep B. YOUNGBLOOD, DIRECTOR, COLLEGE STATION, BRAZOS COUNTY, TEXAS STATION STAFFT ADMINISTRATION B. YOUNGBLOOD, M. S., Ph. D., Director A. B. CONNER, M. S., Vice Director A. H. LEIDIGH, M. S., Assistant Director C. A. FELKER, Chief Clerk A. S. WARE, Secretary M. P. HOLLEMAN, JR., Assistant Chief Clerk J. M. SCHAEDEL, Executive Assistant C. J. GORzYcKI, Technical Assistant VETERINARY SCIENCE *M. FRANcIs, D. V. M., Chief H. SCHMIDT, D. V. S., Veterinarian _ V. J. BRAUNER, D. V. M., Veterinarian CHEMISTRY G. S. FRAPs, Ph. D., Chief; State Chemist S. E. AsBURY, M. S.,_ Assistant Chemist AGRONOMY E. B. REYNOLDS, M. S., Chief G. N. STROMAN, Ph. D., Cotton Breeder PLANT PATHOLOGY AND PHYSIOLOGY J. J. TAUBENI-IAUS, Ph. D., Chief FARM AND RANCH ECONOMICS L. P. GABBARD, M. ., hie V. L. CORY, M. S., Grazing Research Botanist “.44.. H. E. REA, B. S., Assistant SOIL SURVEY T **W. T. CARTER, JR., B. S., Chief ? H. W. HAwKER, Soil Surveyor i EDWARD TEMPLIN, Soil Surveyor W. H. WALKER, Assistant Chemist BOTANY w A. G. PETERSON, B. S., Assistant Chemist H. NEss, M. S., Chief '1 J. E. TEAGUE, B. S., Assistant Chemist if J. K. BLUM, B. S., Assistant Chemist PUBLICATIONS A. D. JAcKsON, Chief HORTICULTURE C. M. MITcHELL, Mailing Clerk A. T. PoTTs, M. S., Chief STATE APICULTURAL RESEARCH ANIMAL INDUSTRY _ LABORATORY . J. M. J0NEs, A. M., Chief H. B. PARKS, B. S., Apiculturist in Charge J. Ldqsn, Ph. D., Animal Husbandman, A. H. ALEX, B. S., Queen Breeder ree ing - G. R. WARREN, B. S., Swine Husbandman MAIN STATION FARM R. M. SHERWOOD, B. S., Poultry Husbandman D. T.KILLOUGH, B. S., Superintendent J. J. HUNT, Wool Grader FEED CONTROL SERVICE ENTOMOLOGY _ B. YOUNGBLOOD, M. S., Ph. D., Director M. C. TANQUARY, Ph. D., Chief; State F. D. FULLER, M. S., Chief Entomologist _ S. D. PEARCE, Secretary H. J. REINHARD, B. S., Entomologist J. H. ROcERs, Inspector C. S. RUDE, B. S., Entomologist W. H. WOOD, Inspector W. O. VIOTOR, JR., Apiary Inspector J. J. KELLY, Inspector J W. R. JORDAN, B. S., Apiary Inspector J. D. PREWIT, B. S., Inspector i H. J. MORRIS, Apiary Inspector D. W. CARLETON, B. S., Inspector ’ SUBSTATIONS N0. 1. Beeville, Bee County _ _ No. l0. College Station, Brazos County E. COWART, 1N1. S., Superintendent (Feeding and Breeding Substation) L. J. McCALL, Superintendent No. 2. Troup, Smith County S. HOTcmuss, Superintendent No. 11. Nacogdoches, Nacogdoches County G. T. McNEss, Superintendent No 3 Angleton, Brazoria County V. HAFNER, B. S., Superintendent . . Beaumont, Jefferson _County A. H. PRINcE, B. S., Superintendent Temple, Bell County A. B. CRON, B. S., Superintendent Denton, Denton County B. DUNKLE, B. S., Superintendent No. 7. Spur, Dickens County _ R. E. DICKSON, B. S., Superintendent No. 8. Lubbock, Lubbock County R. In. KARPER, B. S., Superintendent No. 9. Balmorhea, Reeves County J. J. BAYLEs, B. S., Superintendent A TAs of October 1, 1923. *In cooperation with School of Veterinary Medicine, A. and M. College of Texas. **In cooperation with United States Department of Agriculture. IOn leave. **No. 12. Chillicothe, Hardeman County D. L. JONES, Acting Superintendent No. 14. Sonora, Sutton-Edwards Counties E. M. PETERS, B. S., Superintendent D. H. BENNETT, V. M. D., Veterinarian O. L. CARPENTER, B. S., Shepherd Teachers in the School of Agriculture Carryinlf; Cooperative Projects on the Station .' IS. W. BILsING, Professor of Entomology W. L. STANGEL, Professor of Animal Husbandry, Hogs F. A. BUECHEL, Professor of Agricultural Economics. G. W. ADRIANcE, Associate Professor of Horticulture y W. E. GARNETT. Professor of Rural Sbciolggyf G. P. GROUT, Professor of Dairy Ilusban ry~ CONTENTS. PAGE Part I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '7 Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 10 Part II——The Effect of Individuality on Weight of Fleece . . . . . . . . . 12 Previous Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Method of Handling the Data . . . . . \ . . . . . . . . . . . . . . . . . . . . . . . 12 Individuality in Different Groups of Sheep . . . . . . . . . . . . . . . . . . 23 Individuality at Different Ages . . . . . . . . . . . . . . . . . . . . . . . . . .. 24 The Effect of Season Upon Individuality . . . . . . . . . . . . . . . . . .. 26 The Permanence of Individuality . . . . . . . . . . . . . . . . . . . . . . . . .. 28 The Effect of Barrenness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Individuality and Heredity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 30 Culling for Wool-Production as Compared With-Culling for But- ter-fat-Production and for Egg-Production . . . . . . . . . . . . .. 32 Part III—The Influence of Age and Season on the Weight of Fleece. 34 Part IV——Practical Ways of Culling the Flock for Increased Wool- Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4O SUMMARY 1 1. There is a great deal of opportunity for increasing the average weight of fleece produced by range sheep in Texas. 2. The weight of the fleece produced by a sheep is controlled by three kinds of influences: (a) permanent individual differences between the sheep (called individuality” in this Bulletin), (b) environmental influences which affect some sheep but not others (such as sickness, suckling a lamb, etc.), and (c) environmental forces which affect all the sheep alike (such as age, drouth, etc.). 3. In a flock consisting of 337 grade Rambouillet ewes, 132 grade Rambouillet wethers, 23 registered Rambouillet ewes, and 12 registered Corriedale ewes, kept at Substation No. 14 in Sutton and Edwards Counties, from the summer of 1917 until the summer of 1923, the dif- ferences in the weights of fleeces produced at any one shearing by sheep of the same age and sex were found to be very largely permanent differ- ences, that is to say, due to individuality. The average coefficient of correlation between different fleeces sheared by the same sheep was +607. _ 4:. Individuality was found to be the cause of the following per- centages of the differences in the weights of the fleeces produced by different groups of sheep: 33.5 per cent. for the grade Rambouillet ewes. 38.6 per cent. for the grade Rambouillet wethers. 53.7 per cent. for the registered Rambouillet ewes. 69.2 per cent. for the registered Corriedale ewes. , 5. If all sheep which produced less than the average amount of wool the first time they were sheared had been culled out at the time of that shearing the future average fleece weight of that flock would have been raised about two-thirds of a pound per sheep. 16.2 per cent. of the flock would have been culled when they should have been kept, and 15.2 per cent. would have been kept when they should have been culled as shown by fleece weights at a later shearing. However, most of those which would not have been culled correctly, produced less than a pound either more or less than the average and therefore it would not have made much difference whether they were culled or not. 6. Individuality varies in importance in different groups of sheep, being slightly more important with wethers than with ewes, and being much more important in flocks ofimixed breeding than in flocks of uni- form breeding where the rams have been bought for years with a definite ideal in mind and from only one or two breeders. Culling Will be more immediately helpful on a mixed flock than on a very uniform one. 7. No one age is more accurate than another for culling sheep accord- ing to their fleece weights except that sheep which have been born in a year of drouth can be more accurately culled at twc years of zige than as yearlings. In the interest of economy sheep should be culled as young as possible. Therefore it is recommended that culling be done 6 BULLETIN No. 311. at the yearling shearing except following a drouth, when it should be postponed a year. ‘ 8. One season is as good as another for culling except as noted above for young sheep after a drouth. Fall shearings are about as reliable as twelve-month shearings for culling on the basis of fleece Weight. 9. Consecutive shearings are slightly more .alike than shearings farther apart, but even shearings four and five years apart show a high degree of correlation. 10. When culling ewes two years old or older, one should be careful to observe whether they are dry or are suckling lambs; otherwise, it is possible that culling on the basis of fleece weights might cause more barren ewes to be retained in the flock. 11. Culling will increase the average weightof fleece for the flock of sheep which is culled and it will also increase the wool-producing qualities of the next generation in so far as the individual differences in the sheep culled are hereditary. There are reasons for thinking that very much of this individuality is inherited. 12. The whole matter of individuality may be summed up in the words: “Once a good sheep——always a good sheep; once a poor sheep- always a poor sheep.” There are a few exceptions, of course, but this is much truer than has been generally believed heretofore by the average flockmaster. 13. The wool production of a sheep at one shearing is about as accurate an indication of its future production as one year’s milk and fat record is of the future production of a dairy cow, and is as accurate an indication as the first year’s egg-production is of a hen’s future egg- production. 14. Ewes produce the heaviest fleece at two years of age. Later fleeces are somewhat lighter than the two-year-old, but still heavier than the yearling until old age begins to exert an influence. 15. Wethers produce very much heavier fleeces as tWo-year-olds than as yearlings but their three-year-old fleeces may be still heavier than their two-year-old ones. 16. Wethers produced heavier fleeces than ewes at every age axcept the yearling in this test. * 17. Eight-year-old ewes in the Experiment-Station flock have not yet shown a very decided decrease in wool production due to their age. 18. Abnormal seasons can influence wool production enough to ob- scure the effects of age. 19. Practical methods of culling according to fleece weight are dis- cussed. 20. Length of staple is an indicator of breed at least) but is only fairly accurate. be the heaviest. weight of fleece (within a The longer fleeces tend to \ Bulletin No.’ 311. September, 1923 THE INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON UPON THE WEIGHTS OF FLEECES PRODUCED BY RANGE SHEEP BY JAYVL. LUSH J. M. JoNEs PART 1 INTRODUCTION. Immediately after the establishment of the Ranch Experiment Sta- tion near Sonora in southwest Texas, with suitable arrangements for special range investigations with sheep and goats as Well as cattle, the Texas Station was able to acquire a sufficient number of range-bred Rambouillet ewes to begin a study of several important factors which were thought to have an important bearing on WOOl and mohair in- heritance and production. _ Accordingly in 1917 a comprehensive study of wool- and mohair- production problems was begun by the junior author in his special capa- city as Animal Husbandman in Charge of Breeding Investigations, assisted by E. M. Peters, Superintendent of the Ranch Station. This was the first opportunity that the Texas Station had to systematically accumulate this needed information. In 1921 the senior author, a trained Animal Geneticist, Was employed on the Station Staif, since which time he has assembled and summarized the shearing data so far available at the Ranch Experiment Station. This Bulletin, which represents the accumulation of six years’ shearing records on one phase of wool-production, the effect of age and environ- ment upon the weights of fleeces produced by range sheep, is the first complete bulletin based entirely upon investigations conducted on the Ranch Station, although experiments and observations made there have been used in the preparation of Texas Experiment Station Bulletin No. '29?’ and Circulars Nos. 27 and 28. The importance of Texas as a sheep-raising state is shown b-y the fact that since 1920 Texas has had a larger number of sheep Within its borders than any other state.* However, it is not a matter of pride to learn from the same source that there are a large number of states where the average weight of wool produced by each sheep is greater than in Texas. According to the estimates of the United States Department of Agriculture, supplemented by those of the National Association of Woolen Manufacturers, during the last ten years the number of states *Yearbooks of the U. S. Department of Agriculture. 8 BULLETIN N0. 311. where the average weight of fleece has been as large or larger than in Texas has been as follows: _ Number of states which Year excelled or equalled p Texas 1913 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28 1914 . . . . . . . . . . . . . . . . . . . . . _ . . . . . . . . . . . . . . . . . . . .30 1915 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 1916 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 1917 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 1918 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 1919 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 1920 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1922 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 Of course, these figures are only estimates and there are certain to be some inaccuracies among them, but they are based upon more facts than any other agency has at its command and are the most reliable estimate which we have and clearly point to the fact that the sheepmen of Texas are not getting as many pounds of wool from each sheep as are the sheepmen of many other states, particularly those of the north- ern Rocky Mountain and northern Pacific states. According to these < 57.1) and wrong in less than 28.5 per cent. of the cases. Turning to the actual facts in Figure 2 we find that 41 ewes produced less than the average in 1919 and only nine of those produced more than the average in 1920. Of the 42 ewes which pro- duced more than the average in 1919 only 10 produced less than the average in 1920. Thus there would be only 19 mistakes among the 83 head or only 22.9 per cent. mistakes. Turning to Figure 3, we find that the square of the coefficient of correlation is .432 and we expect to be right in more than 71.6 per cent. of the cases and wrong" in less than 28.4 per cent. As a matter of fact- we find that of the 29 ewes which produced less wool than the average in 1918 only 7 produced more than the average in 1923, and of the 25 which produced more than the average i11 1918 only 6 produced less than the average in 1923. Thus there would have been 13 mistakes among the 53 ewes, or the culling would have been right in 75.9 per cent. of the cases and wrong in 24.1 per cent. Moreover the mistakes which would have been made would have been small ones. Thus in Figure 2 only one of the nine which would have been culled out mistakenly, afterward beat the average by as much as one pound and only four of the ten which would have been kept mis- takenly, afterward produced as much as a pound less than the average. In Figure 3 only two of the seven which would have been culled out mistakenly, afterward beat the average by as much as one pound and only three of the six which would have been kept mistakenly, afterward produced as much as a pound less than the average. Thus it will be seen that the mistakes made in culling on the basis" of fleece weights will be almost entirely among sheep of average merit *Wri_ght, Sewall, 1921. Journal of Agricultural Research 20: 557-585. “Correlation and Causation.” i. ....,....,.~.;¢.im2..-.A£i 18 BULLETIN N0. 311. 4 and that almost all the extremely heavy-shearing sheep will be retainecli and almost all of those producing light fleeces Will be culled out. In the case of Figure 2, if the ewes Which produced fleeces light-erg than the average in 1919 had been culled out, the average Weight of§ fleece produced by the flock in 1920 would have been increased from‘? 9.13 lbs. to 9.73 lbs. In the case of Figure 3 culling in 1918 would have raised the average production of the flock in 1923 from 6.85 lbs”. to 7.54- lbs. Moreover, this culling would not have to be done more thanéf once in the lifetime of each group oi“ sheep but the increased average? production resulting from selling the poor-producers would be obtained§ every year as long as the‘ sheep were kept. Of course, it is advisable to‘; cull all old sheep as fast as they b-ecome ken-mouthed. s The ‘average of all the 161 correlations is +.60 and the square of p that is .368. Therefore we would expect to be correct in more thanf 68.4 per cent. (36.8 + % of 63.2) of the cases if we divided the sheep into; a poor and a good flock equal in size on the basis of one shearing record f and expected each of them to fail into the same two groups at later-ii. shearings. If such a division had been made at the time each group; was first sheared, there would have been 1493 comparisons of two fieecesii from the same sheep. In 512 cases the sheep- was in the poorest half‘! both times, in 241 cases the sheep was in the poorest half of the first‘? shearing and in the best half of the second shearing, in 228 cases the‘? sheep was in the best half at the first shearing and in the poorest halff at the later shearing, and in 512 cases the sheep was in the best half‘, both times. We would have been correct in 68.6 per cent. of the cases; and, as Was pointed out above, in very few of the mistaken cases would the mistake have been more than a pound and therefore it would not matter much whether those sheep were culled out or kept. Moreover“ half of the mistakes would have been in keeping poor producers and most of these mistakes could be culled. out at the next shearing if culling was? continued. Altogether it is impossible to escape the conclusion that? culling on the basis of fleece Weights, would be highly efficient in find-p ing and removing the sheep which consistently produced light fleeces,’- and in thereby increasing the general average yield of wool. in any particular flock. It is of course impossible to discuss each correlation table separately}; but in order to show the variation among these correlations and study?» the possible causes of these variations, a list of the entire 161 coefficients of correlation and their “probable errors”,* together with the number? of sheep included in each is given in Table 1. ' l i‘ a *“Probable error_” is a term used_ by mathematicians to show how likely _it is that the same j results will be obtained if the experiment is repeated under the same conditions. _ For example 5i the coefficient of correlation for Figure-2 is +._655:!;.042. _ This means that if another 835‘ ewes of this same age and breeding had had their fleeces weighed during the same two years the coefficient 0r correlation might not be exactly the same but it would probably not be more i; than +.655+.042 =.697, nor less_ than +.655—.042 =.613._ _If_a coefficient is less than . its probable error it might very easily be the result of chance, _if it is three times its probable; p error the odds are about 22 to 1 that it wasnot an accident, while if it is five times its probable. error the odds are more than 1350 to 1 that it 1S not the result of chance nor an accident. Thus_ in the example used, five times the probable error is .210 and .655 +.210 equals +.865 while , .655——.210 equals +.445. Therefore, the odds are more than 1350 to 1 that if the fleece, weights had been taken on 83 other similar ewes the same years, the resultin coefficient of, correlation would not be less than +.445 or more than +865 and it would e most likely’. to be somewhere between +.600 and +.700. lProbable error is therefore merely a means of, measuring within what limits the facts found in this experiment would repeat themselves iff‘ the experiment were repeated. ' - ‘ INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. Table 1 . A List of All Correlations Between Different Fleece Weights. 19 Number _ of sheep Coefficient of ,1 included correlation Grade Rambouillet_ Ewes, Number 1-100, born mostly in 1915 but some in 1914 and very few in 1916. Long Spring of 1918 with Long Spring of 1919 . . . . . . . . . . . . . . . . . 82 +.714 :1: 036 with Long Spring of 1920 . . . . . . . . . . . . . . . . . 80 +.482 :1: 058 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 37 +.636:1: 066 with Short Spring of 1921 . . . . . . . . . . . . . . . . 36 +.658 :1: 064 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 35 +.546i 080 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 36 +.431 :1: 092 with Short Spring of 1922 . . . . . . . . . . . . . . . . 33 +.401 :1: 099 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 30 +.720:1: 059 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 54 +. 657 :1: 052 Long Spring of 1919 with Long Spring of 1920 . . . . . . . . . . . . . . . . . 83 +.655 :1: 042 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 37 +.688:1: 058 with Short Spring of 1921 . . . . . . . . . . . . . . . . 35 +.772 :1: 046 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 39 +.569 :1: 073 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 40 +.566 :1: 072 with Short Spring of 1922 . . . . . . . . . . . . . . . . 36 +.614 :1: 070 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 29 +.579 :1: 083 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 57 +.705 :1: 045 Long Spring of 1920 'with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 41 +.703 i 053 with Short Spring of 1921 . . . . . . . . . . . . . . . . 40 +.625 :1: 065 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 40 +.546 :1: 087 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 41 +.532 :1: 076 with Short Spring of 1922 . . . . . . . . . . . . . . . . 37 +.400 d: 093 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 34 +.621 :1: 071 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 61 +.611 :1: 054 Fall of 1920 with Short Spring of 1921 . . . . . . . . . . . . . . . . 39 +.753 :1: 047 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 33 +.647 :1: 068 with Long Spring of 1923. . .' . . . . . . . . . . . . . . 32 +.581 :1: 079 Short Spring 0f 1921 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 34 +692 :1: 060 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 30 +.576 :1: 082 Long Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 29 +.779 d: 049 Long Spring of 1921 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . .. 38 +.535 :1: 078 with Short Spring of 1922 . . . . . . . . . . . . . . . . 34 +.520 :1: 084 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 30 +.638 :1: 073 Fall of 1921 with Short Spring of 1922 . . . . . . . . . . . . . . . . 37 +.485 i 085 g with Long Spring of 1923 . . . . . . . . . . . . . . . . . 32 +.679 :1: 064 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 31 +.728i 057 Average for all 36 correlations of ewes 1-100 . . . . . . . . . . . . . + .612 :1: 011 Grade Rambouillet Ewes Nos. 111 to 192, all born in 1918. Long Spring of 1919 with Long Spring of 1920 . . . . . . . . . . . . . . . . . 39 +.416 :1: 089 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 2O +600 :1: 097 with Short Spring of 1921 . . . . . . . . . . . . . . . . 21 +.483 i 113 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 16 +.255:1: 158 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . .. 15 +.165 :1: 169 with Short Spring of 1922 . . . . . . . . . . . . . . . . 14 +.044 :1: 180 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 20 +.758 :1: 064 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 16 +. 693 :1: 088 with Short Spring 0f 1923 . . . . . . . . . . . . . . . . 16 +.382 i 144 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 14 —.074 :1: 179 Long Spring of 1920 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 20 +.793 :1: 056 with Short Spring of 1921 . . . . . . . . . . . . . . . . 21 +.597 d: 095 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 16 +.490:1: 128 with Fall of 1921. a . . . . . . . . . . . . . . . . . . . . .. 15 +.542 :1: 123 with Short Spring of 1922 . . . . . . . . . . . . . . . . 14 +.617 :5 112 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 20 +.717 i 073 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 16 +.642 :1: 099 with Short Spring of 1923 . . . . . . . . . . . . . . . . 16 +.561 :1: 116 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 14 +.715 :1: 088 Fall of 1920 with Short Spring of 1921 . . . . . . . . . . . . . . . . 20 +.573 :1: 101 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 19 +.738 :1; 055 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . .. 15 +.6l7 :1: 108 with Short Spring of 1923 . . . . . . . . . . . . . . . . 15 +.473 :1; 135 20 BULLETIN No. 311. Table 1 .—Continued. A List of All Correlations Between Different Fleece Weights. Number of Sheep Coefficient of included correlation Grade Bambouillet Ewes Nos. 111 to 192, all born in 1918— Continued. _ _ Short Spring of 1921 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 19 +.776:1:.062 wit F 10f 1922 . . . . . . . . . . . . . . . . . . . . . . .. 15 +.568:1:.118 with Short Spring of 1923 . . . . . . . . . . . . . . . . 15 +.600 :1:. 112 Long Spring of 1922 with Fall of 1922 . . . . . . . . I . . . . . . . . . . . . . . . 16 +.682:1:.090 with Short Spring of 1923 . . . . . . . . . . . . . . . . 16 +.563 :1:. 115 Fall of 1922 with Short Spring of 1923 . . . . . 1 . ." . . . . . . . . 16 +.577 :1:. 113 Long Spring of 1921 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 14 +.614:1:.112 with Short Spring of 1922 . . . . . . . . . . . . . . . . 13 +.808:1:.065 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 13 +.264:1:. 174 Fall of 1921 with Short Spring of 1922 . . . . . . . . . . . . . . . . 14 +.330:1:. 161 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 12 +.649 :1:.113 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 12 +.505 :1:. 145 Average for all 35 correlations of ewes 111 to 192. . . . . . . . , . . . . . . . . . . +.535 :1:.023 Grade Rambouillet Ewes Nos. 197 to 297, all born in 1919. Long Spring of 1920 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 29 +.771 :1: 051 with Short Spring of 1921 . . . . . . . . . . . . . . . . 28 +.710 :1: 062 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 28 +.717 :1: 062 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 27 +.660:1: 073 with Short Spring of 1922 . . 1 . . . . . . . . . . . . . 27 +.459 :1: 102 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 28 +.751 :1: 056 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 26 +.391 :1: 112 with Short Spring of 1923 . . . . . . . . . . . . . . . . 26 +4283: 108 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 27 +.618:1: 080 Fall of 1920 with Short Spring of 1921. . . .- . . . . . . . . . . . . 29 +.677:1: 068 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 28 +.820 :1: 042 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 26 +.409 :1: 110 with Short Spring of 1923 . . . . . . . . . . . . . . . . 26 +.642 :1: 078 ' £1921 ‘th L S ' f 1922 . . . . . . . . . . . . . . . .. 28 +.735:1: 059 shm spring o with P31115013??? .9 . . . . . . . . 1 . . . . . . . . . . . . 4 20 + .403 i 104 with Shdrt Spring of 1923 . . . . . . . . . . . . . . . . 26 +.542 :1: 093 ' f 1922 'th F ll f 1922 . . . . . . . . . . . . . . . . . . . . . . .. 27 +.487:1: 099 L°ng Spnng o $5111 siiori) Spring of 1923 . . . . . . . . . . . . . . . . 27 +128 a: 061 Fall of 1922 with Short Spring of 1923 . . . . . 1 . . . . . . . . . . 27 +.450 :1: 104 ' f 1921 'th F ll f 1921 . . . . . . . . . . . . . . . . . . . . 1 . .1 26 +.786:1: 052 Long Spnng o Zvvith Slelori) Spring of 1922 . . . . . . . . . . . . . . . . 26 +.632:1: 079 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 24 +.752 :1: O60 ‘th Sh t S ' f 1922 . . . . . . , . . . . . . . .. 27 +.534:1: 093 Fan of 1921 with 1.0%]; sgirilirilgg g1 192s . . . . . . . . . . . . . . . . . 25 +087 :1: 011 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 25 +.632 :1: 081 Average for all 25 correlations of ewes 197 to 297. . . . . . 1 . . . . 1 . . . . . +.619 :1: 017 Grade Rambouillet Ewes Nos. 298 to 418, all born in 1920. ' ' F ll f 1921 . . . . . . . . . . . . . . . . . . . . . . .. 30 +.491i 094 Long spring of 1921 Slzilori) Spring of 1922 . . . . . . . . . . . . . . . . 30 +.226:1: 117 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 33 +.808 :1: 041 with Fall of 19_22 . . . . . . . . . . . . . . . . . . . . . . . . 31 +.485 :1: 093 with Short Spring of 1923 . . . . . . . . . . . . . , . . 31 +.710 i 060 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 29 +.418 :1: 103 F ll f 1921 w'th Short Spring of 1922 . . . . . . . . . . . . . . . . 33 +5051 087 a ° with Long Spring of 192s . . . . . . . . . . . . . . . . . 30 +508: oss Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . 1 . . . 30 +.626:1: 075 ' 2 ' F ll f 1922 . . . . . . . . . . . . . . . . . . . . . . .. 31 +.755:1: 052 Long Spnng of 192 Slaiori) Spring of 1923 . . . . . . . . . . . . . . . . 31 +.818:1: 040 Fall of 1922 with Short Spring of 1923 . . . . . . . . . . . . . . . . 31 +.653 :1: 070 Average for all 12 correlations of ewes 298-418 . . . . . . . . . . . . . . . . . . . +.589 :1: 033 INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. Table 1.—Continued. A List 0f All Correlations Between Different Fleece Weights. 21 l\/Iurnber of Sheep Coefficient of included correlation Grade Rarnbouillet Ewes, Nos. 420 to 590, all born in 1921. Fall of 1921 with Short Spring of 1922 . . . . . . . . . . . . . . . . . . . . . . . 37 +.622 =1:.068 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 36 +.460 =1=.089 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 3_6 +.646 :1:.066 Long Spring of 1922 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 40 +.385 =1:.091 with Short Spring of 1923 . . . . . . . . . . . . . . . . 39 +. 173 =1:. 105 Fall of 1922 with Short Spring of 1923 . . . . . . . . . . . . . . . . . . . . . . . 42 +.386=1:.089 Average of 6 correlations of ewes 420 to 590 . . . . . . . . . . . . . . . . . . . . . + .445 =1=.044 Grade Rambouillet Wethers, born in 1918. Long Spring of 1919 with Long Spring of 1920 . . . . . . . . . . . . . . . . . 31 +.570 =1: 082 with Fall of 1921 . . . . . . . . . . . . . . . . . . . . . . . . 15 +.275 :1: 161 with Short Spring of 1921 . . . . . . . . . . . . . . . . 15 +.300 =1: 158 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 13 +.754:1= 081 Long Spring of 1920 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 15 +.840 =1: 051 with Short Spring of 1921 . . . . . . . . . . . . . . . . 15 +.740 =1: 079 with Long Spring of 1921. . .' . . . . . . . . . . . . .. 15 +.915=1= 028 Fall of 1920 with Short Spring of 1921 . . . . . . . . . . . . . . . . . . . . . . . 15 +.804=1: 062 Grade Rambouillet Wethers, born in 1919. Long Spring of 1920 with Fall of 1920 . . . . . . . . . . . . . . . . . . . . . . . . 13 +.695 :1=.097 with Short Spring of 1921 . . . . . . . . . . . . . . . . 13 +.356i. 163 with Long Spring of 1921 . . . . . . . . . . . . . . . . . 13 +|.739 =1=.085 Fall of 1920 with Short Spring of 1921 . . . . . . . . . . . . . . . . . . . . . . . 13 +.779 =1=.074 Grade Rambouillet Wethers, born in 1920. Long Spring of 1921 with Fall of 1921 . . . . . . . . . . . . . . . . . .‘ . . . . . . 10 +.787 =1: 081 with Short Spring of 1922 . . . . . . . . . . . . . . . . 9 +.897 =1: 044 with Long Spring of 1922 . . . . . . . . . . . . . . . . . 12 +.555 :1: 134 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 10 +.411 =1: 177 with Short Spring of 1923 . . . . . . . . . . . . . . . . 10 +.768=1= 088 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 8 +.568=1= 162 Fall of 1921 with Short Spring of 1922 . . . . . . . . . . . . . . . . . . . . . . . 10 +.865 =1: 054 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 8 +.795 i 088 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 8 +.660 =1: 135 Long Spring of 1922 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 10 +.726=1= 101 with Short Spring of 1923 . . . . . . . . . . . . . . . . 10 +.643=1= 125 Fall of 1922 with Short Spring of 1923 . . . . . . . . . . . . . . . . . . . . . . . 10 +.477=1= 165 Grade Rarnbouillet Wethers, born in 1921. Fall of 1921 with Short Spring of 1922 . . . . . . . . . . . . . . . . . . . . . . . 25 +.498 :1:. 102 . with Long Spring of 1923 . . . . . . . . . . . . . . . . . 23 +.462 =1:.111 Short Spring of 1922 with Long Spring of 1923 . . . . . . . . . . . . . . . . . 23 +.603 =|=.091 Long Spring of 1922 with Fall of 1922 . . . . . . . . . . . . . . . . . . . . . . . . 23 +.550=1=.098 with Short Spring 0f 1923 . . . . . . . . . . . . . . . . 22 +.161 =1:. 139 Fall of 1922 with Short Spring of 1923 . . . . . . . . . . . . . . . . . . . .; . . 22 +.430 =1:. 112 Average of all 3O correlations involving wethers . . . . . . . . . . . . . . . . . . +.621 3:024 22 i BULLETIN N0. 311. Table 1.——Continued. \ A lgist of All Correlations Between Different Lleece Weights. Number of Shee Coefficient of include correlation Registered Rambouillet Ewes (using only fleeces shorn when the ewes were two years old or older). _ All shearings were long since these ewes were not used 1n the twice-a- year shearing project. 1919 with 1920 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 10 +.7241.101 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9 +.7181.109 1922 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9 +.5341.161 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8 +.3191.219 1920 with 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 +.9851.007 1922 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9 +.8911.046 1923 . . . . . . . . . . . . _ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 +.792 1.089 1921 with 1922 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15 +. 01.051 923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 13 +.7461.083 1922 with 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 +.7821.060 Average of the 10 correlations involving registered Rambouillet ewes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +.733 1.038 Registered Corriedale Ewes (using only fleeces shorn when the ewes were two years old 0r older). _All shearings were long, since these ewes were not used 1n the tw1ce-a- year shearing proJect. 1920 with 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 +.8001.092 1922 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 +. 1 . 140 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 +.7641.106 1921 with 1922 . . . . . . . . . . . . . . . . . . . . ..' . . . . . . . . . . . . . . . . . . . . .. 7 +.8491.071 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 +.9481.026 1922 with 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 +.8561.068 Registered Corriedale Ewes, born in 1921. 1922 with 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 +.9621.025 Average of 7 correlations involving registered Corriedale ewes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . + . 832 1 .026 SUMMARY. No. of Avera e of different the coe icients _ sheep o mvolved correlation Correlations involving grade Rambouillet ewes, 114 tables . . . . . . . 337 +.579 1.010 Correlations involving grade Ramboulllet _wethers, 30 tables. . . . . 132 +.621 1.024 Correlations involving registered Rambouillet ewes, 10 tables . . . . 23 +. 733 1.038 Correlations involving registered Corriedale ewes, 7 tables . . . . . . . 12 . +.832 1.026 All correlations, 161 tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 504 +.607 1.009 (The term “Short Spring,_” as herein used signifies a shearing following a fall shearing. Such shearings represent approximately six and one-half months growth of wool instead of the twelve months growth meant by “Long Spring”) It will be seen from glancing at Table 1 that there is a great variation in the amount of correlation. It is therefore necessary to examine these figures more carefully to determine Whether this variation is accidental, 0r Whether certain groups of sheep, certain ages or certain seasons are characterized by higher correlations than others. Among all the corre- lations only one is negative, and it and two other small positive corre- lations are the only ones which are less than their probable errors. Thir- teen correlations are larger than their probable errors but less than three times as large and therefore We cannot be sure that they are not acci- INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. 23 dental. Thirty correlations are more than three times but less than five times their probable errors and We can be reasonably certain that they surely indicate a positive correlation which would be found if the experi- ment were repeated. Finally, one hundred and fifteen correlations are more than five times heir probable errors and compel us to believe be- yond any reasonable oubt that the weight of fleece produced by a s eep is to a large extent a constant character throughout that sheep's lifeIB INDIVIDUALITY IN DIFFERENT GROUPS. In the search for the causes of the Wide variations shown in the corre- lations one of the most obvious facts is that there is a difference between different groups of sheep. Thus the correlations are remarkably higl; both in the case of the registered Corriedale ewes and in the case of the registered Rambouillet ewes. There are two reasons for this: First, l the registered ewes were all at least two years old when the first shearing used in this test was made and therefore no yearling correlations enter into their averages. The yearling correlations are somewhat lower than the other, as will be shown later. Second, and probably most important, there is probably more hereditary variation among these particular registered sheep than among these grades. The reason for thinking that this is the explanation of the high correlation shown by the Corriedales is that they represent a breed of recent origin which as a whole has not had enough generations of selective breeding to make it as uniform hcreditarily as the older breeds. However, in view of the small num- bcr of Corriedales used in this study it is not safe to draw any general conclusions from it in regard to this breed as a whole. The reason for thinking it of the registered Rambouillets is that they were purchased from different flocks and therefore represent different bloodlines. Also they were selected with the idea of securing representatives of both the B and C types of Rambouillets. Among the different groups of grades the ewes bor11 in 1921 show significantly lower correlations than the others and the ewes born in 1918 show correlations (especially those involving their first fleeces) which are significantly lower than most of the others. The only thing which these two groups of ewes have in common that the other three groups do not, is that they were born in years of severe drouth and were somewhat stunted, those born in 1921 being decidedly stunted. The data on the wethers also show that the correlations involving those born in 1921 are very much below the average for the others, and show very low correlations for those born in 1918 where their first fleece is involved, although their correlations involving later fleeces are very high. This almost perfect agreement between the wethers and ewes born in 1918, and between the wrethers and ewes born in 1921, and be- tween the 1921 and 1918 sheep, when only the correlations involving the first fleeces of the latter are considered, lead us clearly to the con- " clusion that the first shearing of lambs which have been born, suckled, and weaned. in a period of severe drouth is not nearly as reliable an index of their wool-producing ability as are the later shearings. Since the drouth cannot have helped to make the poor-producers better, it must have worked bv preventing the better-producers from developing at an early age to the limits of their individual abilities. In ‘other 24 BULLETIN N0. 311. words the drouths of 191.8 and 1921 and resultant poor grazing have tended to hold all the lambs down to a dead level of mediocre produc- tion. The drouths did not succeed completely in doing this but they did hurt the good producers proportionately more than the poor pro- ducers, as is shown by the correlations which are still positive but are lower than for similar lambs born and reared in years of good grazing. Whether the stunted lambs remain permanently stunted or recover com- pletely during their second and third years is not shown by these figures but they do show that the good individuals recover most or all of their natural advantage over the poor individuals because correlations in‘ volving only shearings later than the first are very much higher than those involving the first shearings. The practical consequence of this is that lambs reared during a drouth should not be culled on the basis of their first-fleece weights or at least should not be culled as severely as lambs reared under conditions of abundant grazing. There is no exact way of correcting for the drouthy conditions under which these 1918 and 1921 lambs were reared but when the effects of these conditions are taken into consideration there seem to be no sig- nificant differences in the importance of individuality among any of the groups of grades unless it be that individuality is a little more important among the wethers than among the ewes. INDIVIDUALITY AT DIFFERENT AGES. It is very important to know whether permanent differences are more easily or surely distinguished at one age than at another, because if that is true all culling can be practiced on sheep as they reach that age with more accurate results than if done earlier or later. To find out whether there is any such influence of age, Table 2 was prepared. In it the correlations are sorted out according to the age of the sheep involved and those pertaining to sheep of the same age are all averaged together. This means that each correlation table is counted in two different places because, of course, it refers to the same sheep at two , different ages. However, since each one is counted twice and only twice, none of the correlations are unduly emphasized by this method. The registered ewes were not included in this table because their correlations were higher and they were not sheared in the fall. There~ fore, it would not be fair to compare a spring shearing in which their correlations were involved with a fall shearing in which they were not. involved. 25 INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eu». cmo. ............................NmTA:wvBM haw. . . . . . . . . . . .. mom. wmm. 2w. www. mum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . 62L wozvm . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . ....-.wQ@Nhw w _ w? _ N. _ Xw w mm _ m. _ w? $03503 macaw wumo? 5 ow< .GOB.M~Q.~.~OU “o QESOE< of 55D om< mo oonoswfl och. éfiichould 03m? . m . . . . . . . . . . . . . . . . . . . . . . hmw. awn. 3%. con. wmm. owv. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . dhziog U . J . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. hmw. wwm. bmw. 1E. . . . . . . . . . . . . . . . . . ‘ . . . . . . . . . loamxomw 33m . . . . . . . . . . ‘ . . . . . . . . .. mmw. 7mm. mmw. an. mwm. ...............I......................QSTwmNwoBM mow. o3». m3. haw. www. www. go. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . $am|>m~ woBM wan. oww. wwm. Nmw. mow. . . . . . . . . . . . . mum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . . . . . flmfll: 33m Amw. . . . . . . . . . . . . . mwm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A57; 38M m5 Hm >5 wmofim Q81 SOQmS. 39H $9 :01“ mm? 89H 5.. ~23“ m5. mgr" 3m. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..§.Ew =< w X» m _ v3 N _ w: _ H _ X wowiufi Qoonw was»? E ow< .GO..5G~O.H~OU we fi5cE< of 55D 33 we ooaonwfi och. .N Q33. 26 " BULLETIN N0. 311. It appears from Table 2 that four and possibly five of the fifteen ages given show significantly lower correlationsthan the others. Those are the six-months shearings, the yearling shearings, the shearing at threeand one-half years, the shearing at five years, and possibly the shearing at six and one-half years, which is significantly lower than the highest of the other ages but not significantly lower than some of the other ages showing high correlations. The low correlations shown by six-months and by yearling fleeces are due to two reasons. First and most important are the drouths of 1918 and 1921, which affected ewes 111 to 192 and 420 to 590 and the wethers born the same years. All of the six-months lamb shearings were made in the fall of 1921 but the yearling~ shearings also included the shearings of ewes 197 to 297, and 298 to 418, and wethers of the same ages which did not suffer the drouths. If due allowancc is made for the effects of the drouths it does not seem that the correlations for these two ages would be significantly lower than for other ages. There is a reason, however, why we might expect the first shearing to be a less reliable indicator of the wool-producing ability of a sheep than other shearings. That is the fact that _the lambs are not all the same age and therefore a good producer dropped late in the season might not show up as well at the first shearing as the poor producer born early in the season, simply because the late-dropped lamb has not had as many days in which to grow its fleece. The fact that the first shearings show up almost as well as the later ones when due allmvance is made for the effects of drouthy years is very encouraging for the practice of culling on the basis of fleece weights because any system of culling which is to be very effective should be carried out when the animals are as young as practical. In the Experiment-Station flock there was .not much difference in the ages of the earliest and latest laorn lambs each year. Our recommendation therefore would be to cull at the yearling shearing except when following periods of extreme drouth, in which case heavy culling should be postponed until the year-and-a-half or two-year-old shearing. We do not know of any satisfactory reason why the correla- tion at the “three-and-one-half; five-, and six-and-one-half-year ages should have been so low. Until we have many more records covering many other sheep in other years we must decide tentatively that no definite age after one year old is better than any other as an indicator of the wool-producing ability of a sheep. The fall shearings are al- most or quite as reliable as the spring shearings. However, to the man not using the scales, differences in the weight of the twelve-month fleece would be much more evident than in the short fleeces. THE EFFECT OF SEASON UPON INDIVIDUALITY. It is theoretically possible that drouth or other unusual seasonal in- fluence might affect the good-producers more than the POOT-PTOClUCGTS or vice versa and therefore that culling might be more successful if practiced in good than in poor years. To determine whether this is so the correlations were classified by years and the result is shown in Table 3. Here, also, each correlation is counted in two different; seasons since-each correlation concerns two different shearings. The correla- tions for the registered ewes are shown but are not included among the averages because they are not represented among the fall and short spring shearings. 27 INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. \' ‘iivl|lolnlti|lil\lllik , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . wQamv Q._N@QFMHOU QMUPW~WQ Os. ..... . . H ......... . . H . . . . s» . ........................................................ . :5 Eaeaemm Wohswwmwm m5. 2%. 2m. Rm .8». S? . . . . . . . . . . . . . . . . A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....E2:@B mmm . OwN . wwm. QNN . fimw . Tvm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . . . . ommlOflfl mvkvm hmm. hNb. ~m©. flab. Nmfi. wNm. . . . . . . . . . . . . . . . . . . . . . . . . . ‘ . . . . . . . . . . . . . . . . . . . . . . . . . .w:»|w®N m®>>m~ Nhw. wmm. Ofifi. fiOh. alum. hww. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kbfllhmw wok/m flaw. . . . . . . . wOb. :3». Owv. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ICNmH\fiHH mvamm .Nw© . . . . Mhw. mNmZ wmm. . . . 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .007; wwkrmm HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .wQ@N-% vllllrllll. l|ll|ll|l4cl|lll J .flO@.wN_®hhOmw 0H3. *0 QNMW 05w CO GOmQQW 05H m0 QOGQEQGm 05b.- .UQUCS.GOQ‘M 0_QN'H. i - - - . - - - . - - . . . - . - . - u - - ~ . . . . > - - - - . - - - . - . . - - - - - . - - - . - - . . - - - . - . . . ‘ I . - - - . . . ' - . . - . - . ~ - - . - - - - - ¢ ~ a ma. ...... . H . . . . . . . . . . . H . . . . Maw. gs. ............ . . ................................... . .5: Qasonamm qfigwwvm ....E. . . . . . . . E ....... .§H ...... ..mSH....HHH..H..HHH...H1H .................................................... ..:2=§> mam“ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .....HHMMMJHMUHUMHHM“MMHMMJMJMMJMMH“Kmmmmmwwmwww . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ‘ . . . . . . . . . . . . . . m®>Pm . ‘ . . . . . . . . . . . . . . . . . . ‘ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . mwam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..OOwlm mvkrmm H~H JH H_H H“ HM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ‘ . . .wnv@mwhw la \ wiwmm fwflq wcvfifi imam oiwflwflhwo wkmwwmwn wfifi mcwhrmmfi wwfi mcfifmwH mfifi wQZoZG Qooam f! . .co$a_o.~.~oU 33 Ho ofim of co common 33 we uocwgfii ash. . 4m Qzfi. 28 / BULLETIN No. 311. l, It is certain from the evidence of this table that season has very little. influence upon the size of the correlation and possibly it has none at all 5 except on the young lambs in time of drouth. There is some indication / . p. that the correlation is highest when wool production. is highest. Thus the correlations are high for the mature sheep in 1919 and for nearly all the sheep in 1920 and for the spring shearing of 1921, which fol- lowed years of abundant grazing. However, the correlations are also high for all except the young sheep in the long spring shearing of 1922, which followed a year of drouth and in which the weight of fleece was less than normal. Hence, this study points to th e conclusion that the amount of correla- tions is almost or quite independent of seasonal influences, being pos- sibly a little higher in years when the average Weight of fleece is high. In other words, there is a possibility that unfavorable conditions may hurt the good-producing sheep proportionately a little more than they do the poor-producing ones, but in general their fleece weights decrease proportionately in bad years and increase proportionately in good years. So far as seasonal conditions are concerned one year is as good as another for culling except, as already stated, in the case of immature sheep reared under conditions of drouth. THE PERMANENCE OF INDIVIDUALITY. Individuality was defined as the permanent characteristics of one sheep, which were diflerent from those of other sheep of the same age, sex, and treatment. However, it might be argued that it is quite pos- sible that some of these differences in wool-producing ability might last over the period of two or three shearings but not last through all the life of a sheep. Thus some sheep might gradually get worse While others improved as wool-producers or a minor ailment might last over at least part of two wool-growing periods,‘ although no sheep were included in these studies which were noticed to be sick or crippled. To answer this important question the correlations ‘were divided according to whether the shearings concerned in each were consecutive shearings or whether there were one or more shearings in between the two shearings concerned in the correlation table. in Table 4. Table 4. ThevEffect of th_e Length of Time Between Shearings Upon the S126 of the Correlation. The results are given Shearings Shearings Shearings Shearings Shearings Consec- separated separated separated separated separated Sheep involved utive by one by two by three by four by five shearings other other other other other shearing shearings shearings shearings shearings . . . . . . . . . . . . . . . .. .661 =!=.0l3 .598=l=.018 .571 =l=.O20 .535=l=.033 .491 51066 .520i.066 Grade ewes 1-100. .659 .591 .595 .579 .609 .657 Grade ewes 111-192. . . 576 .591 .495 . 526 .393 .382 Grade ewes 197-297 . . . 643 .671 .483 . 550 .428 . . . . . . . . . . Grade ewes 298-418 . . .640 .524 .564 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Grade ewes 420-590. . .510 .317 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wethers . . . . . . . . . . . . . .676 .549 .545 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Registered Ramb- ouillet ewes. . . . . .833 .785 .663 .319 . . . . . . . . . . . . . . . . . . . . Re istered Corrie- ale ewes . . . . . . . . . .867 .796 .764 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Number of correla- tion tables included. 64 46 28 14 7 2 INELUENoE OF INDIVIDUALITY, AGE, AND SEASON. 29 The average correlations are given by groups to show that the general trend of the average for all groups is not unduly influenced by any one group. It will be seen that the only significant difference is that be- tween the consecutive shearings and the others. Even this difference is so small that its significance might be doubted were it not for the fact that this same difference is found in all the groups of sheep and in every comparison except between consecutive shearings and shearings separated by one other shearing in the case of grade ewes, Nos. 111 to 192 and Nos. 197 to 297. These figures compel the conclusion that only a very small part of the correlation between the weights of fleeces of the same sheep in different years is caused by the common ailments, pregnancy, the suckling of a lamb, or other temporary conditions affecting some of the indi- viduals and lasting through at least part of two consecutive wool-- growing periods. Using the average figures found for all the sheep included in this study, we find that the differences in the weights of the fleeces shorn at one shearing will be permanent* to the following extent in subsequent shearings: " The next shearing . . . . . . . . . . . . . . . . . . . . .43.’? per cent. The second shearing . . . . . . . . . . . . . . . . . . ..35.8 per cent. The third shearing . . . . . . . . . . . . . . . . . . . ..32.6 per cent. The fourth shearing . . . . . . . . . . . . . . . . .28.6 per cent. The fifth shearing . . . . . . . . . . . . . . . . . . . . ..24.1 per cent. The sixth shearing . . . . . . . . . . . . . . . . . . . ..27.0 per cent. The differences between the last five figures are so small and the numb-er of correlations on which they are based is so small that it is uncertain whether the differences between them are really significant, but it is reasonably certain that the first figure is significantly higher than the others and will probably be found to be higher in future studies made on similar sheep. The registered ewes are not represented at all in the correlations involving shearings separated by four and five other shearings and are only represented by one correlation involving two shearings separated by three other shearings. The absence of the high correlations of the registered sheep from the last three figures given above undoubtedly is one reason why they are lower than the second and third figures and is further evidence that the differences between the last five figures are all probably insignificant. THE EFFECT OF BARRENNESS. The objection has been raised that the high correlation found in this study may be due to the fact that some ewes are permanently barren and ‘therefore produce heavy fleeces while others raise a. lamb every year and therefore produce lighter fleeces. This objection is not well founded because the records show that there was little or no permanent barren- ness and very little temporary barrenness among the ewes of the Exper- inent-Station flock. The effect of pregnancy and of suckling a lamb upon the weight of fleece, has never, so far as the authors are aware, *The square of the coefficient of correlation gives the degree of “determination?” ‘*- 3O BULLETIN N0. 311. been determined exactly. However, it probably does have some in- fluence and, if that influence is very great, culling a band of breeding ewes two years 01d or older on the basis of fleece weights would cause an unusually large proportion of barren ewes to be retained if no atten- tion were paid to this point. It is believed that the proportion of per- manently barren ewes in Texas range flocks is so small that it is of very little practical importance. ' That these high correlations are not due to this cause, however, is shown by the fact that the ‘wethers show a slightly higher correlation than the grade ewes. In fact, ewes which have a lamb some years and are barren other years undoubtedly do more to lower the correlation than the permanently barren ewes and the consistent breeders do to raise it. If these studies were confined to the permanently barren ewes or to the consistent breeders, the importance of the individuality would appear to be even greater than it does with the present figures. Practically, this whole matter of barrenness would have no bearing on culling practiced before the sheep were two years old, but when ewes two years old or older are being culled dry ewes should be required to produce a heavier fleece than ewes with lambs at foot. Probably bar- renness would not make as much difference with the two-year-old evzes as with older ones because two-year-old ewes with lambs would differ from dry ewes of the same age only in the matter of having been preg- nant the preceding fall and winter and having suckled a lamb for a very few Weeks before shearing, whereas with older ewes there would also be a difference in that the barren ones had not suckled a lamb all through the preceding season. INDIVIDUALITY AND HEREDITY. Thus far it has been shown clearly in this Bulletin that there are differences in the weights of the fleeces produced by similar sheep of the same age and sex in the same season and that these differences are, to a large extent, permanent throughout the life of a sheep. On ac- count of this fact culling on the basis of fleece weights has been recom- mended as a means of increasing the average weight of fleece produced by a flock. There is no doubt that it will accomplish that purpose. The question still remains, however, as to whether and to what ex- tent these i dividual differences are inherited. As was stated in the introduction individuality is the combined result of the forces of heredity and “environmental forces which produce permanent effects. We have no convincing evidence which will prove to us that the effects of environmental forces are ever inherited. Hence, we must depend en- tirely upon selecting the results of hereditary forces if we are to produce better and better sheep with each succeeding generation. Therefore, it becomes highly important that we know how much of individuality in this matter of fleece weights is due to heredity and how much due to environment. If it is all. due to environment, culling will be successful in raising the average fleece weight of the sheep on which it is practiced but will not increase the average fleece weight of the next generation. If part of it is due to heredity, culling will increase the average fleece weight of the next generation of lambs also. We do not have conclusive proof to present here of just how important heredity ‘is in the individual differences in the Experiment-Station flock INFLUENCE or INDIVIDUALITY, AGE, AND SEASON. 31 but We do have several indications which make us think that it is very important and enough is known about heredity to indicate "the condi- tions under which it may be more important or less important in privately-ownctl flocks than in the Experiment Station flock. B ‘x Figure 6. Theoretical and actual forms in which data may be grouped in correlation tables. (Solid lines limit the main area in which the data are distributed.) Theoretical grouping when there is no correlation. _ B. Theoretical grouping ‘when there is perfect positive correlation. C. Theoretical grouping when there is a strong positive, but not perfect, correlation due to a large number of common causes acting independently of each other. D. A diagram of the type of grouping actually found in many of the correlation tables discussed in this Bulletin. Presumably caused by a few, very important heredi- tary factors possessed by some but not all of the.sheep in each group. Of course, at bottom, all ivool-production depends upon heredity. More properly speaking, a sheep inherits the ability to produce a. cer- tain amount of WOOl. Whether it actually does produce the full amount, almost that much, or much less, Will depend upon environmental in? fluences. Some characters, such as color and the presence of horns, cannot be influenced very much by environment. Others, such as body size and the amount of fat carried can be very strongly influenced by 32 BULLETIN N0. 311. environment. Probably WOOl-pTOClUClIlOD occupies an intermediate posi- tion, being very largely determined by heredity but subject to a very little permanent modification and considerable temporary modification by environment. It should be kept clearly in mind that the facts and discussions so far have referred to the cZ/ifierenices between individual sheep in the same ~ flock. If the sheep were all absolutely alike in heredity, the differences in fleece weight would be due entirely to environment. If we, could get such a group of sheep we could measure the permanence of their differences and find out how much of individuality is caused by environ- ment. As a matter of fact there are no groups of sheep which are abso- lutely uniform in their heredity, but some flocks are more nearly uni- form than others. For example, the grade flock in this experiment was descended entirely ‘from one band of ewes even for several generations before they were purchased by the Experiment Station and carewas used in selecting the rams even though they varied from light B to heavy O types. The registered Rambouillet ewes, on the other hand, were pur- chased from different sources and represented different lines of breed- ing and different types. Presumably the higher correlations and more permanent individual differences of the registered ewes are largely due to their greater differences in heredity. Even in the Experiment Station’s grade flock, hereditary differences. are probably very important because there is a difference in type among the grades and type is believed from practical experience to be largely determined by heredity and closely associated with the weight of fleece. Also in the correlation tables worked out to determine the importance of individuality there were too many individuals which were very poor each time or very good each time; that is, too many near the extreme upper left-hand corner or near the extreme lower right-hand corner of the tables to be caused by a large number of different environmental forces. The most reasonable explanation is that they differed from each other in a very small number of very important hereditary factors affecting wool-production. Figure 6 illustrates this point. The practical consequences of this are: first, that culling will im~ prove the wool-production of succeeding generations of sheep even in as uniform a flock as the Experiment Station grade flock; second, that. flocks more uniform in type and heredity than this one will not be benefited as much either in the present generation or in the succeeding generations; third, that flocks less uniform in type and breeding than this one will be benefited more by culling, both in the increased pro- duction of the present generation and in the increased production of‘ succeeding generations. Therefore the hereditary benefits to be derived from culling are going to depend entirely, and the present benefits will depend somewhat on the amount of hereditary variation in wool-produc- tion in the flock which is to be culled. CULLING FOR WOOL AS COMPARED WITH CULLING FOR BUTTERFAT AND~ EGG PRODUCTION. Culling dairy cows on the basis of their record for one year or even- on the basis of their record for seven-day or thirty-day periods has been: advocated for many years and has fully justified itself when put im INFLUENCE 0F INDIVIDUALITY, AGE, AND SEASON. 33 practice. Hence it is interesting to compare the reliability of culling dairy cattle on the basis of their milk production with the reliability of culling sheep on the basis of their wool production. The correlation between different advanced registry records for 365 days made by the same cows has been found’; to be +.692 for the Guernsey breed and +567 for the Holstein breed, and +535 for records made by a single large Jersey herd extending over a period of more than twenty years. When these correlations are compared with the +.5'79 obtained for all the grade Rambouillet ewes in this study it will be seen that the weight of the fleece of a sheep at a single shearing is practically as reliable an index for culling that sheep as is an entire year’s milk record for culling a dairy cow. Probably it is even more reliablebecause advanced reg- istry records are selected data,—i. e., cows have had to produce at least as much as the minimum entrance requirements two different times in order to be included in that study. Moreover, nearly "every cow was under the care of the same owner during both tests but different owners gave widely different care, whereas this flock of grade Rambouillet ewes was under the same care all the time. The comparison with the large Jersey herd is fairer for both these reasons but even it has an advantage over the sheep records in that some ewes would occasionally lose their lambs and therefore would not be giving milk during part of the time they were growing wool. The records of the cows are of course not complicated by an irregularity of this kind except the number of days they were carrying a calf. Therefore, it is no exaggeration to say that culling for wool-production can be fully as successful as culling for milk-production has been. Culling hens on the basis of their trapnest records and on the basis of certain body characteristics has also been recommended very widely. The first-year and second-year egg-production records of white leghorn hens have been found* to show a correlation of between +.539 and +.554. The body characters which show the largest correlation with egg-production are the color of shanks and beak (in the light colored breeds)**. Color of ‘shanks has shown a correlation of +.622, the lighter color indicating the larger egg-production, and color of beak has shown a correlation of +503. Other characters have shown. smaller correlations. Hence it will be seen that the fleece weight at one shear- ing is as reliable an index for future wool-production as the egg- production for one year is for future egg-production or as any external body character is for egg-production. ‘lfiowen, John W. and Marie S., 1922. Annual Report of the Maine Agricultural Experiment Station, page 33. *Harris, J. Arthur and Lewis, H. R., 1922. Genetics, Vol. 7, pages 274-318. _ “Sherwood, R. M., 1922. Texas Agricultural Experiment Station, Bulletin 295. Correla- tion between External Body Characters and Annual Egg-Production 1n White Leghorn Fowls. 34 BULLETIN N0. 311. PART III THE INFLUENCE OF AGE AND SEASON ON THE WEIGHT OF FLEECE. It was possible to study individuality without taking either the effects of age or season into consideration, merely by using i11 any one correla- tion table only sheep of the same age and sex Which were sheared the same two seasons. It is not, however, possible to study the effects of age and season separately. This is obviously true because if shearing records of the same sheep at different ages are used they will also be influenced by differences in season While if we compare sheep of differ- ent ages the same year their individual wool-producing abilities may be different. It is possible, however, to study the influepce of age andysonrand if the study is carried on witi enough sheep over ,,~’a great many years, “accurate knowledgemcanyplqmg psecurerlflwofcjhefeirgt effect of age andiseasonalchanges, The facts reportediin this Bulletin are not sufficient to answer all t e questions about the effect of age and season on fleece weight, but they tell something about it and, as such, are worth presenting. l It is a matter of common knowledge among wool-growers that the average Weight of fleece changes with age and is different in different years, but, so far as the authors know, there has been published only o e study of the amount of these changes*. In that study the con- , lusion was drawn that the lightest fleece was the yearling fleece and the heaviest was that shorn at three years of age. However, it was observed that in some crosses the second fleece was the heaviest, and it could only be said of one age that its fleece was surely either smaller or larger than the others. That was the yearling fleece which was sig- nificantly smaller than the others in almost every comparison. The sheep used in the Arizona study were Hampshire, Shropshire, Tunis, native and various crosses between them. It does not necessarily follow that the Rambouillet will behave in the same way with respect to age as the four kinds of sheep named »above. Moreover, the grade Rambouillets in this study were bred under differ- ent conditions and the records were taken in different years from those in the Arizona study. It will be found that these results disagree With the Arizona results chiefly in that the three-year-old shearing is not the heaviest, except in the case of the wethers, and that later mature shearings can be even lighter than the yearling shearings in very un- favorable years. The combined influence of age and season upon the Weight of fleece y produced by the different groups of grade Rambouillets in the Experi- ment Station flock is shown in Tables 5 and 6 and Figures '7 and 8. *Ariz0na Agricultural Experiment Station, Twenty-seventh Annual Report, 1916, pp. 281-283 . INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. . 35 Table 5. The Influence of Age and Season on the Weight of Twelve-month Fleeces. (Expressed in Percentage of the Yearling Fleece Weight.) Kind of Sheep \ 1918 1919 1920 1921 1922 1923 Ewes ~1—100 . . . . . . . . . . . . . . . . . . .. *100.0 109.3 111.7 101.9 92.3 81.7 Ewes 111—192 . . . . . . . . . . . . . . . . . . . . . . . . . . .. 100.0 116.8 104.9 101.4 86.4 Ewes 197-297 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 100.0 117.8 100.6 86.1 Ewes 298-418 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 100.0 99.9 86.2 Ewes 420-590, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100.0 **94.5 Wethers born in 1918 . . . . . . . . . . . . . . . . . . . . . 100.0 127.2 141.4 . . . . . . . . . . . . . . . . Wethers born in 1919 . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 100.0 134.0 . . . . . . . . . . . . . . .. Wethers born in 1920 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100.0 132.2 116.6 Wethers born in 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100.0 **127.5 *Most of these ewes were three years old at this shearing. They were not purchased until the summer_ of 1917, when most of them were two years old but some of them were three and a very few were only one. Therefore, 100% represents about a three-year-old fleece in this case instead of a yearling fleece as in all other cases. **Comparison between different individuals, since yearling and two-year-old twelve-months fleeces were not taken from the same individuals. These sheep were divided in the fall of 1921 into two equal groups for alternate once-a-year and twice-a-year shearings. The dIVISIOD. was made by running them through a cuttin chute and taking alternate individuals for the two lots. Therefore, the division was probab y as nearly equal as it was possible to make 1t Table 6. The Influence of Age and Season on the Weight of Short Fleeces. (Expressed in Percentage of the Yearling Fleece Weights.) Fall, Spring, Fall, Spring, Fall, Spring, 1923 Kind of Sheep 1920 1921 1921 1922 1922 Ewes 1-100, born mostly in 1915. 55 1 8.9 45 4 60.3 . . . . . . . . . . . . . . . . Ewes 111-192, born in 1918........ 55 0 64.9 47 4 58.2 40.2 49.1 Ewes 197-297, born in 1919........ 60.7 5.4 49.1 61.4 42.7 48.7 Ewes298—418, born in 1920........ . . . . . . . . . . . . . . .. 52.4 57.2 44.1 50.2 Ewes 420—590, born in 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48.3 59.3 Wethers born in 1918 . . . . . . . . . . . . . 67.8 84.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wethers born in 1919 . . . . . . . . . . . . . 68.5 82.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wethers born in 1920 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64.1 79.1 59.5 75 .2 Wethers born in 1921 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58.6 80.1 (Fleeces of ewes 1-100 expressed in percentage of the three-year-old fleeces, since the yearling fleece weights were not available.) Table No. 6 shows that the fleece Weights were rather light in the fall of 1921 and spring of 1922 and very light in the fall of 1922 and spring of 1923. This agrees very Well with Table No. 5. Table No. 6 also shows that the spring fleeces were distinctly heavier than the fall fleeces. The fall of 1920 and fall of 192]. clips include about five and oiie~half months’ WOOl. in each, the fall of 1922 includes about five months’, the spring of 1921 includes six and one-half months’, the spring of 1922 includes seven months’, and the spring of 1923 includes only six months’ growth of Wool. A comparison of Tables 5 and 6 will show that the combined weight of the two short clips is slightly greater than that of the single long clip, usually between five and ten per cent. greater. This twice-a-year shearing experiment is not yet completed. however, and there are many other things besides the weight of the Wool to be considered. These other things will be discussed in detail in the bulletin which will be published when the experiment is completed. Therefore .no further comment will be made on this subject at this time except to say that the results so far are not very favorable to the prac- tice of shearing twice a year. 36 i BULLETIN N0. 311. 13o g 12° $1‘ . 11o $- 100 g- eo 743- 80%- EWOS N05. 1 '- 100 - —- —- -— Ewes Eos. 111 - 192 - - - - — — — - - Iiwes Nos. 197 - 297 *~+~—~—~—*—'-*-—~ Ewes non. 29s - 41a Ewes Nos. 420 - 590 193s 19.19 19,20 19,21 19,22 19,25 Fi ure 7. The Influence of Age and Season on the Weight of Twelve-month Fleeces Pro- duce by Grade Ramboulllet Ewes. _ The weight of the first fleece is taken as _one hundred per cent. and later fleeces are compared directly with the first one. The first one 1s the yearling fleece 1n every case except_w1th ewes Nos. 1-100, most of which were three years old 1n 1918. The figure shows graphically that the two-year-old fleece tends to be the heaviest but that the influence of season 1s greater than that of age. (The data are taken from Table No. 5.) 1407;- 1307;- 120%- 110$‘ 100%- —i—-——-——— wethers born in 1918 — - — — — — - -- wethera born in 1919 -.—-— ——— ——- ‘Iethsrs born in 1920 -— o —— a —— wethers born in 1921 1911 9 IQPO 1 9'21 1 9.22 19t23 | Figure 8. The Influence of Age and Season on the Weight of Twelve-month Fleeces Pro- duced by_Grade Rambouillet Wethers. _ The weight of the first fleece is taken as one hundred per cent and later fleeces are compared directly with the first one. The first one is the yearling fleece in every case. The figure shows graphically that the two-year-old fleece is very much heavier than the yearlin fleece. Probably in mosfnormal years the older fleeces are still heavier than the two-year-ol fleece. Seasonal influences do not affect the wool-production of the wethers as much as they do that of the ewes. INFLUENCE or INDIVIDUALITY, AGE, AND SEASON. 37 In order to eliminate the differences in individuality the average weight of the first fleeces shorn was taken as one hundred per cent. and later fleece Weights were compared With this first Weight. This first fleece Weight was the yearling Weight in every case except for grade ewes 1-100, for Which the yearling fleece Weight Was not known, since the most of them Were three years old when they Were first sheared as Experi ment Station property. The comparison is between the same individuals in every case. Thus, with ewes 111-192 there were 39 head which Were sheared both in the spring of 1919 and the spring of 1920 and the comparison between those years is based on the average fleece Weights of the entire 39 head. In the spring of 1921, however, only 16 of the 39 head produced full year’s fleeces (most of the others having been sheared in the fall of 1920 as part of the experiment on shearing tWice-a-year, and a few having died) and therefore the average fleece weight of these 16 sheep in the spring of 1921 is compared directly with the average fleece weight of the same 16 sheep in 1919 instead of with the average of the entire 39 in 1919. In this way differences due to individuality are eliminated and it makes no difference Whether the sheep which died during the period covered by these records were the best or the poorest. There is an exception to this in the case of the lambs born in 1921, for half of them Were sheared as fall lambs and hence they have never yet all produced a full year’s fleece at the same time. If these lambs “Here fairly divided the first fall, the comparison is all right; but if one group happened to get better individuals than the other, the comparison is not so reliable. Since the division Was as fair as was possible in the judg- ment of the men dividing them and there were 22 to 25 Wethers in each group and 35 to 40 ewes in each group, it is probable that the differ- ences between the two groups were very slight and that the comparison of their 1922 and 1923 records is as reliable as the other comparisons. There were slight differences in the date of shearing in difierent years with the following result: ' The 1918 fleece represents about eleven months’ WOOl. 1919 fleece represents about twelve months’ Wool. 1920 fleece represents about eleven and one-half months’ Wool. 1921 fleece represents about twelve months’ Wool. 1922 fleece represents about twelve and one-half months’ wool. 1923 fleece represents about eleven months’ wool. Therefore the figures for 1918 and 1923 are a little lower and those for 1922 are a little higher than they should be. The only other very marked difference between seasons was in the amount of rainfall and the resultant scarcity or abundance of grazing. The rainfall for the twelve months preceding each shearing (May 1 to April-30) is as follows: 1919 shearing . . . . . . . . . . . . . . . . . . . . . . . . . H2443 inches 1920 shearing . . . . . . . . . . . . . . . . . . . . . . . . . . .28.65 inches . 1921 shearing . . . . . . . . . . . . . . . . . . . . . . . . .28.73 inches 1922 shearing . . . . . . . . . . . . . . . . . . . . . . . . . ..18.71 inches 1923 shearing . . . . . . . . . . . . . . . . . . . . . . . . . ..29.21 inches 38 BULLETIN No. 311. The rainfall records were not taken preceding the 1918 shearing at the Experiment Station itself but it is known that that season was very dry and was preceded by a very dry one, and the records for the corre— sponding period at San Angelo, one hundred miles farther north where the rainfall is usually a little less than at the Experiment Station, show 8.16 inches of rainfall for those twelve months. Of course, the distribution of the rainfall is fully as important as its amount and when that is considered it is evident that there was a short drouth in the summer months of 1918, and that the severe 1921 drouth began about the middle of June and lasted until the following March, there being only one month between these two dates when there was as much as one inch of rainfall. In summary, then, the shearings of 1918 and 1922 followed periods of severe drouth- and scanty grazing, but the other shearings probably were not affected. (The 1919 shearing may have been somewhat affected); The pastures were decidedly understocked with sheep the first four years. Turning to Table No. 5 one will see that the variation in the amount of wool produced from year to year is not large. The average weight of fleece was small. in 1922 and quite small in 1923. For the two groups of older ewes it was also small in 1921. Even when it is re- membered that the 1922 shearing represents twelve and one-half months’ growth of wool whereas the 1923 shearing represents only eleven months’ growth, there still seems to be a need for further ex- planation of the low average weight of the fleeces in 1923. We know that they were unusually light in their shrinkage and can offer tentatively only the explanation that the unusually numerous and heavy rains in the three months preceding shearing may have washed out much more of the dirt and grease than usual and that the early shearing before hot weather began would have caused much less yolk to be present than if the sheep had been sheared a month later. However, it may be that a small part of this decreased production in 1923 is merely an after effect of the drouth of 1921 as a. result of which the sheep started into the season of 1922 under-nourished and in poor condition. We need more data on the subject of how long it takes a sheep to recover from the under-nourishment experienced during a severe drouth, and we need much more data on the variations in shrinkage percentages of wool grown under different conditions and clipped at different seasons. Such data on shrinkage are being collected at present by the wool-scouring laboratory of the Experiment Station not only for the Experiment- Station flocks but also for the flocks of many cooperating wool-growers of the state. On account of the fact that the fleece weights were abnormally low in 1922 and 1923 it is not possible from this study to establish exactly A the effect of age upon fleece weight. However, it seems clear that the two-year-old fleece is decidedly heavier than the yearling fleece and that with the ewes the two-year-old-fleece is heavier than the subsequent ones. With the wethers it is probable that the three-year-old fleeces may be. even heavier than the two-year-old fleeces, as they were in one of the cases shown in Table 5. This difference in the effect of age on ewes and wethers is probably due to two things: first, the aged wethers are larger than the aged ewes and, second, the ewes drop their first lambs at two years of age and their fleeces shorn at that time show only the INFLUENCE or INDIVIDUALITY, AGE, AND SEASON. 39 effects of pregnancy while later fleeces show the effects both of pregnancy and of having suckled a lamb through the preceding season. These data do not show how old a ewe must be before her wool pro- duction decreases on account of old age. Ewes 1-100 Were mostly eight years old in 1923 and many of them were showing outward signs of age; yet their average fleece Weight decreased that year and the preceding year only a little more than the fleece weights of the other groups of ewes. Admitting that the data are not complete enough to prove beyond all reasonable doubt the conclusions given below, nevertheless We think it wise to draw from these data and from practical experience the following tentative conclusions with respect to the effect of age on fleece weight :\ 1. The fleeces of two-year-old ewes are heavier than those of yearling) ewes; usually ten to twenty per cent. heavier in normal years. 2. The two-year-old is the heaviest fleece produced by breeding ewes.‘ a Later fleeces are somewhat lighter than those produced by two-year-old ewes but are heavier than the yearling fleece. 3. Two-yiear-old wether fleeces are very much heavier than yearling wether fleeces, usually twenty to forty per cent. heavier. 4. Aged wethers produce fleeces still heavier than two-year-old wethers. 5. Wethers produce heavier fleeces than ewes at every age except the yearling. l 6. The influence of very drouthy seasons is greater upon the fleeces of the ewes than upon those of the wethers and is great enough to cover up all the influences of age upon the fleeces of the ewes. '7. Fleece weights of ewes do not decrease very much on account of old age before they reach the age of at least seven or eight years. 40 BULLETIN No. 311. PART IV PRACTICAL WAYS OF CULLIN G THE FLOCK FOR INCREASED WOOL PRODUCTION. As a means of attaining a higher degree of perfection in the breeding of sheep in Texas, wool-growers are urged to adopt a systematic method of culling 0r removing all animals that do not come up to the average standard of the flock. For example, all off-type sheep including those possessing weak constitutions, as Well as those possessing light frowzy fleeces should be removed from the flock in order to provide accommoda- tions for more productive and profitable ones. Culling has been practiced among the leading pastoralists of some of the more progressive sheep countries for many years and it is no doubt true that the leading position in the production of some of the choicest fleeces in the world, which is today held by the Australians, has been due to the rigid method of culling and classing their flocks, although, of course, the general use of sires of superior breeding has had its Wholesome effect. The information presented in this Bulletin serves as a further illus- tration that there is likely to be a Wide amount of variation in the amount of Wool produced by the various individuals of the average flocks of this state. Furthermore, it shows that the general tendency under similar conditions and management is for a sheep that shears a heavy fleece one season to produce a heavy fleece at subsequent shearings. Like- wise under similar conditions a sheep which shears a light fleece at one shearing will be very likely to produce a light fleece at all future shearings. However, it is clearly recognized that this study deals only With weight of fleece and that this is only one of the factors which determine how much profit can be obtained from keeping each sheep for a year. Culling on the basis of fleece Weights alone Would therefore be a very imperfect method of flock improvement. The scales tell nothing but Weight and the man Who would cull and breed according to Weight of fleece alone would be pursuing just as foolish a course as the dairy- man who culls and breeds his cows solely on the basis of the amount of milk they produce without regard to its quality and Without taking into consideration the body characteristics of the cows. Such a prac- tice Would be better than making no effort at all to improve the flock and it would be fairly certain to increase the quantity of wool produced, but it Would also be very likely to lead frequently to the keeping of sheep weak in constitution and undesirable in mutton conformation, or pro- ducers of wool too coarse or too straight, or containing kemp or “beard” hair, or some other undesirable kind of wool. All these things should be considered in culling. Australian wool growers who are not qualified to class their own flocks employ sheep classing experts who are recognized as being thoroughly eflicient and must be familiar with the various kinds of wool produced by the flocks on which they Work and also with the class of wool in greatest demand. Sheep classing requires not only that the. classers have a knowledge of wool but also that they be fairly good judges of sheep. INFLUENCE or INDIVIDUALITY, AGE, AND SEASON. 41 Since we are not so fortunate as to have expert sheep classers avail- able for such a service in Texas at the present time it is necessary to Igapproach the culling problem from another angle. It has been one of fjthe aims and purposes of the Texas Station through the medium of the fWool Grading and Scouring Plant to grade and scour representative lsamples of wool for the wool growers of the state with a view of placing sin their hands some valuable information with reference to the qhality d condition of samples scoured. The majority of Texas wool growers we fairly competent to distinguish between a good mutton sheep of de- Egsirable type possessing a strong constitution and one which is lacking i" constitution and development. Therefore, with his knowledge of the t that it would be utterly futile for him to attempt to produce the ghest class of wool on a sheep that does not have a strong constitution, will be possible for the grower, with the aid of information upon the bject of wool, to make some material progress in the direction of flock provement. The amateur wool grower. should first of all establish his ideal and nstantly strive toward that goal. If he is not competent to dis- guish between the best fleeces carried by certain members of the ck and those produced by some of the mediocre animals he should k the advice of someone who is qualified to select his best sheep, both Lm the standpoint of type, and wool-production. This will enable u to establish a definite type toward which he can well afford to work the improvement of his flock. Sheep of the fine-wool breeds having the following faults should be led from the flock: ' 1 All sheep that are undersized or possess weak constitutions. i2. Sheep that are off type, that is, those possessing long legs out of .portion to their size, weak backs, over-shot or under-shot jaws, etc. .3. Sheep producing light frowzy wool lacking in density. f4. Sheep producing wool with too much variation in the size of fibers. .5. Fine-wool sheep producing patches of black wool on any part of ,6 body. ‘6 All those producing kempy hairs: Such hairs are ‘likely to be d about the face, forelegs, and thighs. Kemp is a separate and _.= 'ct fiber from the so-called “beard” hair which is frequently found . the folds or breech. This “beard” hair, as it is sometimes designated, i also very objectionable. {Constitution and type should be given the first consideration by the yders who are attempting to raise the standards of their flocks. In selection of the sires and dams or in the culling of the off-type indi- uals from the flocks, breeders should bear in mind that there is a "n standard of excellence which should be maintained for the ewes gwell as for the rams. = ram should show masculinitywhile the ewe should display a fair of refinement without any tendency toward weakness. The ram 111d be masculine, commanding, and active. The head should be well "on, being neither too large nor too small, but proportionate to the of the body. The forehead should be slightly arched and broad een the eyes; eyes should be bright and prominent with determined 7» ession. The fae should be of moderate length with a full slightly ,ded nose, covere ‘with fine silky hair; muzzle well developed, nos- 0 _ any black streaks. '4 VPURITY—N o hair, kemp, or black or dark brown fibers. Cut heavi- 42 BULLETIN N0. 311. trils open, wide and thick; neck of medium size, well rounded muscular, tapering gradually from the head to the shoulders. w chest should be broad and roomy with a well developed heartgi withers broad or rounded, top-line straight without any deficiency mediately back of the Withers; back strong, ribs well sprung, hips developed and not too prominent. The rump should be modera long, broad, and not too drooping; tail well set on. Hind qua ii should be well rounded and muscular to hock; legs of medium len straight, wide apart, and well set under body. The hoofs. of the fi wool breeds should be of a clear amber orhoney color and free f " The foregoing description could in a general way be appliedf, representative breeding ewes. However, breeders recognize that m, hers of the ewe flock should be more highly developed in the hind q i ters than are the rams Which have more strongly developed fronts. i - It is impossible in the limited space allotted in this Bulletin to e. into a lengthy discussion of the desirable and undesirable characterise,’ ' of wool, therefore the authors have deemed it advisable to reprod" the Wool Score Card which Was prepared several years ago by '. fessors Bray and Hill of the Colorado Agricultural College and s‘ University of Wyoming, respectively. Although this card has not b‘ oflicially adopted by animal husbandmen it has been carefully prepa’ and should prove valuable to not only practical sheepmen but to mal husbandry Workers and students as well. THE BRAY AND HILL WOOL SCORE CARD. QUALITY on FINENESS——FiI1e fiber, breed or grade considered. Not‘ a mixture of fine and coarse fibers. Not a Wide difference be- ' tween shoulder and breech. Fine areas large, coarser ones small LENGTH—Should be clearly of combing length for the grade, i. e., fine, 2%”; 1%- blood, 3"; % blood, 3%”; i blood, 4". Lengths more than 5 inch greater of no additional value except in wool coarser than 71* blood. Fibers that lie together, all the same length. Little variation over main parts of the fleece. A minimum of short wool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. SoUNnNEss—Strong throughout, no weak spots. Fibers of uni- form thickness from base to tip except for tapering of yearling fleeces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ily for: beard-hairs on wrinkles of fine wooled sheep, dark fibers mixed through the fleece of downs, and coarse hair on "the breech of crossbreds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. CHARACTER- (a) Fiber, evenly crimped throughout, crimp close and dis- tinct, fibers parallel except for sufficient binders to hold the fleece together. Tips free from Wastiness. No frowzy wool . . . . . . - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (b) Soft and springy to touch, elastic under pressure . . . . . .. COLOR———White, bright. Main fleece free from stains. Minimum of stained areas around breech and on belly . . . . . . . . . . . . . . . . INFLUENCE 0F INDIVIDUALITY, AGE, AND SEASON. 43 CONDITION-- (a) Yolk, moderate in quantity, light color evenly distributed o (b) Free as possible from natural adhering sand and dust, and from heavy tags and sweat locks . . . . . . . . . . = . . . . . 3 (c) Free from burrs, chaff, seeds or other foreign matter, not i mouldy or mothy, free from excessive paint, not tied with sisal or rough jute twine or with excessive amounts of twine. Not cotted . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 20 Total . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 With the assistance of this card the experienced wool-grower who has a practical knowledge of judging sheep will be able, with a. little practice, to distinguish between those individuals producing wool of high quality and character and those that produce open frowzy fleeces. It will be an easy matter to identify these individuals producing wool with a wide variation in the degree of fineness from shoulder to breech, as a result of a careful examination of the fleece. Kemp and beard hair i.f present on certain individuals can be easily detected. According to American wool manufacturers one of the outstanding criticisms of Texas fine wool is that it is sent to market in rather uneven lengths. Therefore, it will serve as a reminder to Texas wool- growers that the length of the WOOl fiber should be given greater con- sideration in order that a larger amount of fine combing wool, which should be at least 2% inches in length, might be made available for the manufacturer by the Texas growers. Wool-growers are engaged in the sheep business as an enterprise and are anxious to make maximum profits. They should therefore strive to produce wool that will meet with the requirements of the leading manufacturers of the country. Upon length of staple depends the strength of the yarn to be spun, there being much less waste in the spinning of a long than of a short fiber. Moreover, long—sta.pled fleeces tend to be heavier than short-stapled ones, at least as between fleeces of the same degree of fineness. (This is shown by a correlation of —|—.56 on 69 Rambouillet fleeces at the Wyoming Station; an average correlation of +35 on 259 grade Ram- bouillets in nine groups in unpublished studies by the Texas Station; certain experiments of the United States Bureau of Animal Industry reported by Marshall which show the same thing but are not figured in terms of correlation. These are correlations of length and weight taken on the same fleece, and the length of one fleece and the weight of subse- quent fleeces probably would not show as high a correlation as this). Therefore, it seems that the breeder who selects for length of fibers is also selecting to some extent for weight of fleece, but this point needs further study. Fineness of fiber is necessary in order that the spin- ners may be enabled to spin the yarn to the requisite number of counts. Fineness to a certain degree is an indicator of breeding but there is a posibility of such a condition as over-fineness of fiber. Luster is one of the striking features of crossbred and braid wools. Whiteness is an indicator of good breeding. A yolk of a light orange color is not objectionable; however, different colors of yolk on the same sheep are not desirable. Members of the flock which carry fleeces with an overabundance of thick gummy and clotted yolk of a dark color i should be discriminated against since it is difiicult to remove such s _ represents the typical merino fiber. They give greater elasticity to I 44 BULLETIN N0. 311. from WOOl during the scouring process. . All flock-masters have observed that there is a considerable varia in the amount of crimp in the wool fibers of the different types » fleeces represented. Some members of the flock produce wool in w g the crimp is almost imperceptible, while others produce wool having‘ distinct crimp which almost forms a half circle either way. Such cri fiber and improve the general felting properties. At shearing time if the wool grower is desirous of culling the li shearers and off-type individuals from his flock he can proceed v- efficiently if he establishes the practice of separating the sheep by ~a into several groups. For instance, if he should decide to cull t: yearling flock he will find the culling operation to be much easierj he will first separate his yearlings from the remainder of the a, before he proceeds with the culling operation. He should first deci on the percentage of ewes to be culled; then he should select seve typical representatives of the best as well as several of the most _ desirable members of the flock in order that he may establish a mo effective culling basis upon which to proceed more readily and accuratel * If the grower feels qualified to rely on his judging ability exclusive in the culling of the flock he can proceed to examine each individ carefully just before shearing time. He should place a mark or bra on the face of each sheep that he plans to cull, while those that are r5 be retained can go unmarked. The paint mark placed on the cul if located on the face 0r ear will probably not remain visible more th _ a few days; therefore, after the ewes have been sheared they shoul either be separated from the main flock or given a body paint brand c» order that their identity may be known as long as they are retained i the flock. If the grower does not feel competent to rely on his judgment ext clusively he can secure a small amount of additional help at shearin time and weigh the individual fleeces. This will give him accurate formation with reference to the amount of wool produced; however, i was mentioned above, growers are advised against the establishment 0- the custom or practice of culling based exclusively on shearing weights; i For example, it is possible that a sheep possessing a very weak consti-* tution might produce an extra heavy or attractive fleece, and since it; is the aim of every progressive breeder to eliminate off-type, under-sized animals from the flock, he should adhere to this practice even though an occasional off-type flock member here and there should produce a! heavy fleece. Furthermore, the scales do not reveal such deficiencies- or objections as the presence of an undue amount of kemp and the so-i: called “beard” hairs so frequently found on certain sheep. ‘It is there-Q,‘ fore obvious that if the xvoolqgrower hopes to make a fair‘ degree of progress in the direction of flock improvement he must learn some of _» the rudiments of wool judging and classing since he cannot hope to .f attain the highest possible degree of perfection in his flock Without first being able to distinguish between superior and inferior fleeces. a J INFLUENCE OF INDIVIDUALITY, AGE, AND SEASON. 45 LITERATURE CITED. wen, John W. and Marie S. 1922. Annual Report of the Maine ii Agricultural Experiment Station, pages 21-60, especially page rris, J. Arthur and Lewis, H. R. 1922. Genetics, Vol. 7, pp. 274- 318. “The Correlation Between First and Second Year Egg Pro- _ duction in the Domestic Fowl.” _l, J. A. 1921. Wyoming Agricultural Experiment Station, Bulle- tin No. 127. “Studies in the Variation and Correlation of Fleeces from Range Sheep.” _ a J. A. National Wool Grower, Vol. II, No. 4, p. 17. ’ 1.921. hall, F. R. 1918. American Sheep Breeder, Vol. 38, No. 2, pp. 92-3. “Experiments in Breeding Range Sheep.” 1 wood, R. M. 1922. Texas Agricultural Experiment Station, Bul- _i letin No. 295. “Correlation Between External Body Characters ; and Annual Egg Production in White Leghorn Fowls.” ht, ‘Sewall. 1920. Proceedings of the National Academy oi Sciences, Vol. 6, No. 6, pp. 320-332. “The Relative Importance of gHeredity and Environment in Determining the Piebald Pattern 50f Guinea-pigs.” p ht, Sewall. 1921. Journal of Agricultural Research, Vol. 20, all To. 7, pp. 557-585. “Correlation and Causation.” s, R. H. and Cunningham, W. S. 1916. Annual Report Ari- fzona Experiment Station, pp. 281-288. ‘ - rvwWwwqnwIj'v'flflf\ini§ F!‘ an =v I’