QH 431 C265 GENETIC STUDIES OF RABBITS AND RATS BY W. E. CASTLE PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON WASHINGTON, MAY, 1922 ^GENETIC STUDIES OF RABBITS AND RATS BY W. E. CASTLE \> PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON WASHINGTON, MAY, 1922 BL* CARNEGIE INSTITUTION OF WASHINGTON PUBLICATION No. 320 Copies of this book first issued MAY 2 21922 TECHNICAL PRESS WASHINGTON, D. C. GENETIC STUDIES OF RABBITS AND RATS. PART I. SIZE-INHERITANCE IN RABBIT CROSSES. In the last fifteen years numerous studies have been made of the inheritance of characters which are quantitatively variable or fluc- tuating. As a result of these studies it has become clear that in many cases fluctuation is due to non-genetic causes, to the environ- ment rather than to the constitution of the germ-cells. In such cases selection is without effect in modifying the racial character. This is the accepted explanation of the negative results obtained by Johannsen in selecting beans for increased or decreased size, and of the similar results of Ewing in selecting plant-lice for altered body dimensions, and those of Jennings and others with paramecium. But in a majority of cases variation due to genetic causes occurs in association with that due to non-genetic or environmental causes; in fact, it is possible to distinguish between the two only by the results of systematic selection. When the environment is kept constant and a race does not change in response to selection, we assume that no genetic variation is present. But if the race does change in response to selection, we have no alternative but to assume that the variation is genetic in character. Body-size in birds and mammals shows well the simultaneous yet distinct action of genetic and non-genetic agencies. The amount and quality of the food supplied to an animal limits its size, yet if food is supplied in abun- dance and of proper quality, some races of animals attain greater size than others. This is the result of genetic differences. The analysis of such genetic differences is difficult. A pioneer attempt was made by Galton (1889) in his study of human stature, the inheritance of which he characterized as blending. This term was adopted by Castle et al. (1909) in describing the inheritance of ear-length and body-size in rabbits. A Mendelian interpretation of size-inheritance was later advocated by Lang (1910), based on the multiple-factor hypothesis of Nilsson-Ehle and this has now received general acceptance. Davenport (1917) has recently applied it to human stature, but has gone a step farther in assuming that the genetic factors which govern size in one part of the body are often not the same as those which govern the size of other parts. Punnett and Bailey (1914, 1918) in their studies of size-inheritance in poultry and in rabbits bring forward a different hypothesis. 4 -KTIC STUDIES OF RABBITS AND RATS. They attempt to explain the results of size-crosses with the aid of genetic factors affecting the size of all parts, but relatively few in number, a different potential effect on size being assigned to each of the assumed factors. With a view to throwing further light, if possible, on this confused situation, a renewed investigation of size-inheritance in rabbits was undertaken by me in January, 1917. To clarify the matter, it seemed desirable to obtain races as pure as possible and as unlike as possible in size, to breed each separately and study its variability and at the same time and under similar conditions to cross the same races and study the variability of the FI and F 2 generations of off- spring. For this purpose two small races of rabbits were selected, Polish and Himalayan, and one large race, that of the Flemish Giant. Stock was obtained from exhibitors at the last previous Boston show who had been awarded prizes for the excellence of their animals in these several breeds. In carrying out the original plan to study size-variation in the uncrossed races, I have been only partiallysuccessful, but a good series of cross-breds has been secured and studied. In the pure races (plate 1) the number of adult individuals studied was, Polish 23, Himalayan 8, Flemish 5. Adult cross-breds have been produced as follows: Polish X Flemish, F, 27, F 2 137; Polish X Himalayan, F, 25, F, 55; Himalayan X Flemish, F! 17, F 2 70. The total number of adult individuals studied for size-characters in this investigation is 367. The characters studied are weight, ear-length, and the fol- lowing bone measurements: (1) skull length, (2) skull width anterior to orbit, (3) skull width posterior to orbit, (4) length of femur, (5) length of tibia, (6) length of humerus. Weights were taken at intervals of about two weeks, from the time of weaning to the attainment of full growth. As animals are apt to increase in weight through the accumulation of fat after the cessation of general growth, the adult weight of an animal is taken as the maximum weight attained under one year of age. Small races of rabbits attain maturity earlier than large ones. For example, Polish rabbits are usually full-grown at ten months of age, whereas Flemish may grow slowly after they are one year old. The skeletal measurements studied are based for the most part on animals which had attained the age of fourteen months, but some measurements have been used of younger animals, when it has been found by com- parison of different measurements that full growth had probably been attained. Adult ear-length is attained much earlier than full body-size. Growth of ears is practically completed at twenty weeks of age. Accordingly, it has been possible to use the ear- measurements of animals which died before they attained full body-size, and so the totals for ear-length studies are greater than those for weight GENETIC STUDIES OF RABBITS AND RATS. O (which is supposed to be complete at one year) or for bone- dimensions, which as a rule are based on specimens fourteen months old. GROWTH-CURVES BASED ON WEIGHT. Most of the rabbits were born in the spring and summer months, May to August inclusive, but some were born as late in the year as November and a few in the early part of December. None was born in the months January to March inclusive. The young were usually weaned when one month old, at which time periodical weighing of the animals was begun. It was considered very desirable to raise all the individuals born, for fear that if small-sized individuals should succumb in competition with their larger brothers and sisters, the statistical conclusions might be vitiated thereby. To this end, free use was made of foster-mothers. Young under a week old may readily be transferred from the nest of one mother to that of another. In only one case have I known a mother to refuse to care for foster- children substituted for her own of like age. Of course, notwith- standing all our precautions, many young rabbits died before reaching maturity, either from intestinal troubles in hot weather or from nasal troubles ("snuffles") in cold weather. But a careful study of our records indicates that there was no differential selection in favor either of large or of small rabbits in these deaths. The young rabbits were supplied with an abundance of suitable food, in summer grass and oats, in winter hay, oats, and cabbage or carrots. Occasionally a rabbit was not weaned promptly at one month of age. If, for example, an individual seemed feebler than its fellows at weaning- time, it might be left with the mother a week or two longer. Again, differences in size were often observed at weaning-time between litters of like ancestry, but of unlike number in individuals. If a doe nurses six young, they will average smaller at weaning than will young of similar ancestry nursed by two mothers, each of which divides her milk among three young. Fanciers believe that this affects the ultimate size of the offspring. They think that the largest rabbits are obtained by rearing only two or three young to a litter and by allowing these to nurse the mother for six weeks or two months. Our observations do not support this view, but indicate that the ultimate size attained is influenced very little, if at all, by the size of the young at weaning, provided they are in good health and growing condition and are thereafter fed abundantly. Figure 1 illustrates the matter well. Rabbits 2650 and 2834 were half-brothers. They had the same pure Flemish father and their mothers were of the same pure Polish stock. Rabbit 2650 was weaned at 29 days of age and grew slowly until he was 71 days old, when he weighed 600 grams. Rabbit 2834 was raised by a foster- a GENETIC STUDIES OF RABBITS AND RATS. mother alone, receiving all her milk until he was weaned at 71 days old, when he weighed 1,530 grams, more than 2.5 times as much as his half-brother at the same age. Notwithstanding this handicap in his favor, he was overtaken in size by his early- weaned half- brother at age 194 days, and was surpassed by him in adult weight by about 150 grams. The bone dimensions of the two rabbits were very similar, with a slight but probably not significant superiority in favor of the late-weaned individual. In the case of neither rabbit does the growth-curve show any interruption of healthy growth. The form of the growth-curve is affected by the superior nutrition of 2834 during his first 2 months, but the ultimate size attained is not influenced thereby. Figure 1 also shows the growth-curve of a third rabbit (2578), half-brother to each of the others (2650 and 2834). He was born 160 210 340 270 AGE. IN DAYS Fio. 1. Growth-curves in weight of three F t rabbits which were half-brothers, all having the same pure Flemish sire but each having a different pure Polish mother. Male 2578 wms born July 26, 1917; male 2660 was born August 9, 1917; male 2834 was born Decem- ber 7, 1917. Note that the rabbit which was largest as an adult (2650) was smallest up to 160 days of age, whereas the rabbit (2834) which was much the heaviest as a young rabbit because he received all the milk of a foster-mother up to weaning at 71 day. of age, nevertheless fell below 2650 in adult weight. just two weeks earlier than 2650, was weaned at the same age (29 days), but was slightly heavier at weaning and held the lead up to age 144 days, when he was practically full-grown. His ears were then of maximum length, 10.4 centimeters. On the same day [December 17) the younger rabbit (2650), although lighter in GENETIC STUDIES OF RABBITS AND RATS. 7 weight, had longer ears (10.9 centimeters), a fact which foreshadows his greater adult size. He kept on growing after passing age 144 days and ultimately became much heavier. His bone-measurements were also greater, except in one instance, femur-length, which was practically the same. In the case of this pair of half-brothers, the environmental conditions were substantially identical. They were born within a fortnight of each other, weaned at the same age, and grew up under the same seasonal conditions on similar food, yet they diverged in adult weight, bone-measurements, and ear-length more than the pair (2650 and 2834) which had such different environ- mental conditions. From facts such as these, it is believed that accidental differences in environment (though we have striven to avoid them) have little to do with the size-differences observed among our rabbits. On the other hand, it is clear that our purest races of rabbits are not perfectly homogeneous genetically as regards size, since with the most carefully controlled environment size-differences occur. Figure 2 shows the growth-curves of two female rabbits, 3099 and 3100, litter mates borne by the same pure Himalayan mother to the 20 ts 1 S-o 2 t-" is - T- 99 J 1 .. 100 T ^ ^ ; . ft #' i - r > FIG. 2. Growth-curves of two FI rabbits, litter-mates borne by the same pure Himalayan mother to the same pure Polish sire, and kept together in the same pen throughout their growing-period. Note difference in character of the growth-curves. The rabbit which was larger at weaning, matures earlier and remains smaller as an adult. 10 20 Z70 300 same pure Polish sire. Both were weaned on the same day at the age 28 days. They were kept together in the same pen during subsequent growth. At birth, 3099 was the larger of the two sisters and maintained her superiority for about four months, when her growth-rate began to slow up, while that of her sister, 3100, kept on steadily. At about the age 170 days the originally smaller rabbit surpassed her sister in size and held this relative position thereafter. In every bone measurement she was a little larger than her sister, but in ear-length they differed little, the smaller rabbit being credited with having slightly longer ears. From the weight-records made for each rabbit it is possible to construct a growth-curve, as in the cases just discussed. Such curves have been plotted for each pure-bred and each FI cross-bred s GENETIC STUDIES OF RABBITS AND RATS. rabbit raised. From these curves, readings have been taken giving the approximate weight of the individual at the ages of 30, 40, 60 days, and so on, at 30-day intervals, to the age 360 days. Combining the weight records for each group of individuals, an average growth- curve is obtained for the group, as shown in figure 3. In this figure it is indicated that Polish rabbits are small at 30 days of age, but grow with fair rapidity until about 180 days old. Then the growth-rate declines and growth is practically ended at the age 210 days. In many cases the weight actually decreases again after attaining a maximum at sexual maturity, or shortly thereafter, as Punnett (1918) has observed. But such decline is not invariable, and it may result from a variety of causes, such as less nutritious FIQ. 3. Growth-curves of the three races of rabbits, Polish, Himalayan, and Flemish, and of the groups of FI hy- brids obtained by crossing the three pure races. The average adult weight of each group of Fj hybrids is shown for comparison with FI, be- low which it falls in every case. AGE. IN A ft, IN UAY9 food, a cold, or lactation (in females) . When normal conditions are restored, the weight usually rises again and subsequent accumulations of fat usually carry the final weight above the puberty maximum. In plotting the growth-curves the maximum weight attained in the first year is considered as persisting thereafter. This will account for the fact that the plotted average curves do not show any decline, even though the weight-curves of individuals would in many cases do so. The rabbits that show increases in weight, due largely to fattening subsequent to puberty (180 to 210 days), will explain the fact that the weight-curve continues to rise until the end of the period plotted, 360 days. In the case of Himalayan and Flemish rabbits, the weight at the age 30 days is greater than that of Polish rabbits. The growth-rate uso greater, so that the growth-curves continue to diverge more and more, one from another. The slowing up of growth at puberty GENETIC STUDIES OF RABBITS AND RATS. 9 (180 to 210 days) is also less abrupt and the increase in weight sub- sequent to puberty is greater. The average weight at 360 days of Polish rabbits of both sexes is less than 1,400 grams, for Himalayan rabbits it is 1,800 grams, and for Flemish it is over 3,200 grams. In Polish rabbits, as compared with other and larger breeds, the initial weight is less, the growth-rate less, and the completion of growth comes earlier. All these phases of growth combine to make the ultimate weight smaller. Comparison of the growth-curves of Polish and Flemish brings this point out very clearly. The Flemish curve is far above the Polish curve at the outset and diverges from it increasingly up to the age 360 days. The Himalayan growth curve lies everywhere between the Polish and Flemish curves, but is much nearer to the Polish. In form, however, it is more like the Flemish curve, in that growth continues longer after puberty. FI rabbits produced by crossing Polish with Himalayan have a greater initial weight than either pure parent race and grow faster. At puberty (180 to 210 days) the FI rabbits surpass Polish by over 500 grams. They are more than 40 per cent heavier. As compared with Himalayans of like age, they are more than 20 per cent heavier. At age 360 days, the FI rabbits are 45 per cent heavier than Polish, but only 9 per cent heavier than Himalayan, since the slower-growing Himalayans have lessened the superiority which the FI rabbits showed at puberty, but have not extinguished it. F 2 rabbits from this cross, obtained by mating the FI rabbits, brother with sister or half-sister, show an average adult weight of 1,632 grams, which is intermediate between the adult weight of Polish and Himalayan rabbits, the races originally crossed, being 20 per cent heavier than Polish and 10 per cent lighter than Hima- layan. The superiority in size of the FI animals is a purely FI phenomenon, as we shall see. It never persists into the F 2 generation. Its existence is nevertheless an important practical consideration, which makes grading and the crossing of breeds highly advantageous under certain circumstances. In the cross of Polish with Flemish rabbits, FI is in size well above the intermediate between the parent races, another illustration of FI vigor, although in this case FI does not surpass the larger race in size. The form of the growth-curve is also intermediate. Maturity is attained early, as in the Polish race, but the practical cessation of growth comes later than in the Polish race, at 240 days rather than at 200. The adult weight of the FI rabbits, at the age 360 days, averaged 2,539 grams, as compared with 3,240 for Flemish and 1,359 for Polish. The intermediate would be 2,300 grams, which the FI group surpasses by more than 10 per cent. In the Himalayan- Polish cross, FI surpassed the intermediate by over 20 per cent, and F 8 surpassed it by 3 per cent. 10 GENETIC STUDIES OF RABBITS AND RATS. In the Flemish-Polish cross, as in the Himalayan-Polish cross, F falls below F, in size. It is found to be 2,128 grams as compared with 2,539 for Fi. It is even less than the intermediate, 2,300 grams, by about 8 per cent. In the Flemish-Himalayan cross, the growth-curve for F! is very similar in form to the curve for pure Flemish. There is a gradual falling in the growth-rate at about 210 days, which indicates the usual slowing-up influence of puberty, coming about a month later than in Polish rabbits. But growth continues even after this retard- ing influence sets in, just as it does in Flemish rabbits. The weight at 360 days is 2,781 grams as compared with a strict intermediate between the parent races of 2,523 grams, which it surpasses by 10 per cent, about the same relation found in the Polish-Flemish cross. Again F shows a falling off as compared with FI to 2,466 grams, which is 2 per cent less than the intermediate. To summarize the foregoing observations: (1) In races of small-sized rabbits the initial weight is small, the growth-rate is low, and growth terminates early. Conversely, rabbits of large-sized races have a large initial weight and growth energy, which not only makes them grow faster, but also makes them grow a longer time than do rabbits of small-sized races. (2) When races of small rabbits are crossed with a race of large rabbits, the initial weight, growth energy, and duration of growth are all intermediate in character, but are greater than the strict intermediate in the Fi generation and approximate it or fall slightly below it in Fj. (3) When two races of rabbits (such as Polish and Himalayan) are crossed, which do not differ greatly in size, FI may surpass either parental race in vigor of growth, though not in duration of growth. Consequently the maximum advantage in size of FI over the parent races will be attained at puberty. F will approximate the interme- diate in size between the parent races originally crossed. WEIGHT AND SEX. In the foregoing pages the average weight of groups of individuals has been under discussion without reference to individual variation within each group or the effect of crossing on such variation, or the relation of weight to sex. In any discussion of the inheritance of weight, these questions must be taken into consideration. First let us consider the relation of weight to sex. The "standards" of the various breeds of rabbits formulated by breeders in England and the United States make no specification as to the relative sizes of the two sexes in the small breeds of rabbits, but in the large breeds they regularly specify a larger size for females than for males. Thus, the English standard for the variety Steel Gray Flemish Giant GENETIC STUDIES OF RABBITS AND RATS. 11 (plate 1, F) specifies " Bucks shall be not less than 11 pounds and does not less than 13 pounds." An American publication gives the standard weight for Japanese rabbits as "bucks 7 pounds, does 8 pounds," that of French Silvers as "bucks, 9 pounds; does, 10 pounds or over." The existence of such specifications in breed standards implies that, at least in large races of rabbits, the female either is naturally larger than the male of like genetic constitution, except as to sex, or else that the breeders have some expectation of making it so. The relative weight of the two sexes in the groups of rabbits which I have studied is shown in table 2. In the Polish breed, males were found to be slightly heavier than females. The single male Himalayan studied differed very little from the average of the 5 females; but in the cross-bred rabbits of all three combinations, viz, Polish-Hima- layan, Polish-Flemish, and Himalayan-Flemish, females were heavier by from 50 to 200 grams (2.5 to 7.5 per cent) than were males derived from the same cross. It seems clear, therefore, that the breed stan- dards are correct in recognizing the greater weight of females than of males, at least in the heavier breeds of rabbits. The difference is a natural one, not arbitrary, but the amount of the difference has probably been exaggerated in the standards, which call for a difference of about 15 per cent. This is double the greatest difference observed by me. INHERITANCE OF WEIGHT. Individual variation in weight can methods. Maximal first-year weight the few cases of rabbits received as adults. For the present, the slight difference in size of the two sexes will be disregarded. The general character of the variation in weight is presented graphically in figures 4 to 6. The data on which these graphs are based will be found in table 3. Figure 4 shows that the modal weight of Polish rabbits is about 1,400 grams, while Himalayan rab- bits are about 400 or 500 grams heavier. The Fi cross-breds are heavier than either pure race, as shown also in the growth-curves (fig. 3) . They vary symmetrically around the class whose center is 1,949 grams, designated class 19 in figure 4. The Fj generation, which contained just best be studied by statistical will be dealt with, except in NO. F^.PXH F..PXH H nri P^3 10 II 12 13 UVI5 16 17 18 I9ZOZI ZZ 23 24 WEIGHT, IN HEKTOGRAMS Fia. 4. Polygons showing variation in weight of pure Polish (P) and pure Himalayan (H) rabbits, and of their Fi and F t hybrid offspring. twice as many individuals as 12 GENETIC STUDIES OF RABBITS AND RATS. Fi, varies about class 16 as a mode, being about 300 grams lighter in average weight than the Fi groups. The variability of this group as measured by the standard deviation is only slightly greater than that of FI, being 233 grams as compared with 218 grams for Fi. 20- 14- rt- F,,PXF ..EL 1 12 13 * IS 16 n 18 19 ffl 21 ZL Z3 24 Z526 Zl X 23 3031 3233 34353637 3839 4041 WEIGHT, IN HEKTOGRAMS Fio. 5. Polygons showing variation in weight of pure Polish (P) and pure Flemish (F) rabbits, and of their FI and F 2 hybrid offspring. Figure 5 shows the effect upon weight of the widest of the three crosses, that between Polish and Flemish. The parental groups are separated by over twenty classes. F! lies midway between them, its mean falling in class 25. The variation polygon is rather flat and wide, which would seem to indicate lack of complete genetic uniformity in one or both parent races. FI is of similar form, but is shifted about four classes farther to the left. A sim- ilar movement of the mean occurred in the Polish-Himalayan cross between the FI and the F 2 genera- - tions. The variability, as J^jn^ r measured by the stand- _ c Mm _ ar d deviation, has in- creased by about 25 per cent, from 198 grams in to 257 grams in F 2 . WEIGHT, IN HEKTOGRAMS Fio. 6. Polygons showing variation in weight of pure Himalayan (If) and pure Flemish (F) rabbits, and of their F, and F, hybrid offspring. Figure 6 shows for the Himalayan-Flemish cross the variability in weight of F! and F,. F! is strictly intermediate between the parental races. Its mode lies in class 27, but the average is somewhat higher, being 2,827 grams (table 3). The mode of F, lies three classes GENETIC STUDIES OF RABBITS AND RATS. 13 lower in class 24, the mean being 2,472 or about three and a half classes lower than FI. The variability of F 2 is about 40 per cent greater than that of Fi, being 230 grams instead of 162. In neither of the Flemish crosses does the F 2 variation extend up- ward into the range of pure Flemish, but is separated from it by from four to seven vacant classes. In other words, the larger parental type is not recovered in F 2 , doubtless because the number of inde- pendent genetic factors involved is too great. If the larger parental type is not recovered, it is not to be expected that the smaller type would be recovered, since this would involve a chance recombination of factors equally improbable of realization in a limited number of offspring; but in reality the F 2 range does in every case extend down- ward into the range of the smaller parent type; therefore either the smaller type is more readily recovered, or what appears to be the smaller type recovered in F 2 is really not such, but is a new genetic combination or combinations which resemble in gross weight the original parental type. The latter alternative seems more probable. This view is supported by the observations on ear-length, which character is closely correlated with weight, and yet in regard to which neither the small. nor the large parental type is recovered in F 2 , all F 2 variates occurring in the intermediate region. EAR-LENGTH. Ear-length is a character easier to study than body-weight, because the adult condition is attained earlier and the races studied are more sharply distinguished as to ear-length than as to weight. The ears, as a rule, have attained their full growth at the age 150 days, or even earlier in the case of the Polish race, whereas the weight continues to increase slowly after that age. The ear-length was recorded at the same time that the rabbits were weighed, though less frequently, as little change was noted in the ear-length after the age 150 days. After a rabbit had attained that age his ears were measured merely often enough to make sure that no further change had occurred and that the earlier measure- ments had been accurate. A variation of 1 or 2 millimeters was frequently noted in the measurements, and the final rating of each rabbit was accordingly based on the approximate mean of the recorded measurements. The measurement taken was read from a ruler placed between the base of the right ear and the head, the ear being then held vertically against the ruler and slightly stretched and the reading made at the ear-tip. Small races of rabbits have short ears; large races have both longer ears and longer skulls, as we shall see. The variation in ear-length of the groups of rabbits studied is shown graphically in figure 7. The Polish rabbits studied range 14 GENETIC STUDIES OF RABBITS AND RATS. in ear-length from 81 to 88 millimeters, Himalayan from 92 to 97 millimeters or about 1 centimeter longer-eared than Polish, while the three Flemish individuals used as parents, notwithstanding their diversities in weight, are remarkably uniform in ear-length, ranging from 143 to 147 millimeters, or about 6 centimeters greater than Polish and 5 centimeters greater than Himalayan. The three races are very distinct and each by itself very uniform. The data on the ear-length of the cross-bred rabbits is summarized in table 4 and is shown graphically in figure 7. CAR LENGTH, IN CENTIMETERS Fio. 7. Polygons showing variation in ear-length of the three pure races of rabbits and of their Fi and Fj hybrid offspring. In the cross between the Polish and Himalayan breeds, which differ in ear-length by about 10 millimeters, long ear appears to be completely dominant. Variation is about the mode of the Himalayan race and the average is practically the same (94.8 in the Himalayan, 94.9 in the FI group). But in F, the variation is much greater and the range extends downward so as to include the mode of the Polish race as well as upward to cover the entire range of the Himalayan race and of FI. The cross of Himalayan with Flemish gives results which are easier to interpret. These two races considered singly are very uniform in ear-length. Each varies closely about a mode widely separated from that of the other race (see fig. 7). F! is about 4 or 5 millimeters below the intermediate and varies little. F 2 has the same relation to the intermediate, with practically the same mean, but is much more variable, the standard deviation (table 4) having risen from )8 millimeters in F, to 5.85 millimeters in F 2 , and the range from 8 millimeters to 24 millimeters. Nevertheless, the range of F 2 does not GENETIC STUDIES OF RABBITS AND RATS. 15 extend into the range of either parental race. A gap of 8 millimeters separates the shortest-eared F 2 from the longest-eared Himalayan rabbit, and a gap of 13 millimeters separates the longest-eared F 2 from the shortest-eared Flemish. Hence it is clear that neither parental combination is recovered in the total of 70 F 2 individuals studied. Clearly, many independent factors must differentiate the parent races as regards ear-length. The cross of Polish with Flemish is a still wider one than that just described and Fi is somewhat less uniform. It is, indeed, about as variable as the F 2 obtained from the Himalayan-Flemish cross, which fact indicates a greater genetic variability in the Polish than in the Himalayan race as regards ear-length. Again, in this cross FI falls below the intermediate between the means of the parent races by about 5 millimeters. F 2 shows greatly increased variability as compared with FI. The range has risen from 14 to 30 millimeters, and the standard deviation from 3.76 to 6.05 millimeters (see table 4). Nevertheless, the range of F 2 does not extend into that of either parental race. Four vacant classes separate it from the range of pure Polish and 20 vacant classes separate it from the range of pure Flemish. It accordingly approaches the smaller race more nearly than the large one (as in weight), but neither parental condition reappears in F 2 , although this consists of 131 individuals. Again we are forced to conclude that . numerous independent genetic factors affecting ear-length differentiate the parent races. The question may be raised whether there is a difference between the sexes as regards ear-length, as there appears to be in regard to weight. Table 5 answers this question in the negative. The sexes differ very little in ear-length and neither is consistently greater than the other throughout the groups. BONE MEASUREMENTS. The length of the skull has been measured, in these studies, from the notch in the ventral margin of the occipital foramen to the anterior surface of the incisor teeth in a straight line, measurement being made with a caliper rule and readings taken in tenths of milli- meters. Differences in skull-length between the races of rabbits studied are less in amount than those which distinguish these same races as regards ear-length. But it is desirable to ascertain whether skull-length differences are inherited in a similar way and whether they are correlated with differences in ear-length and body-weight. Hence the study of skull-length will have value, even though its results are not as clear-cut as those derived from other studies. Polish rabbits have a skull-length of from 63 to 69 millimeters. Himalayan rabbits have somewhat longer skulls, ranging from 66 16 GENETIC STUDIES OF RABBITS AND RATS. to 72.5 millimeters. The skull-lengths of 5 Flemish rabbits studied range from 82.5 to 88 millimeters (see table 6). The cross between Polish and Himalayan rabbits produced off- spring surpassing both parent races in skull-length, as in body- weight, a manifestation, no doubt, of heterosis or cross-bred vigor. The FI averaged 70.2 millimeters in skull-length, the larger parent race averaging 68.9 millimeters, the intermediate between the parents being 67.3 millimeters. The average skull-length of F 2 in this cross was exactly equal to that of the larger parent. It would seem that the pure Polish and Himalayan races may have been below their genetic possibilities in skull-length, owing perhaps to inbreeding. The Himalayan-Flemish cross (table 6, F,, H.XF.) produced F t animals averaging a little larger than the intermediate between the parent races, and F t animals averaging a little less. But the F 2 animals were more variable than the F! animals, their standard deviation in skull-length being 2.85 millimeters as compared with 1.79 millimeters for FI. The Polish-Flemish cross produced an Fj generation close to the intermediate between the parent races as regards skull-length, but FI fell more than 2 millimeters below the intermediate. Again Fj was more variable than FI, the standard deviations being 3.15 and 1.18 respectively. Tables 7 and 8 show the variation of the several groups of rabbits in skull-width measurements taken anteriorly and posteriorly respec- tively to the orbit. It appears that Polish rabbits, though weighing less than Himalayan rabbits and having shorter skulls, nevertheless have skulls slightly broader. The F! rabbits from the Polish-Hima- layan cross have skulls broader than those of either parent (as well as longer, see table 6), and this superiority is retained in part in the FI generation. These statements apply both to the anterior and to the posterior skull-width measurements. Although the Polish rabbits have slightly broader skulls than the Himalayans, neverthe- less they do not transmit as much skull-width in crosses with Flemish as do the Himalayans, for in every case the skull-width of the Hima- layan-Flemish cross-breds is slightly greater than that of the Polish-Flemish cross-breds. The explanation probably is that the Himalayans transmit greater general body-size (weight) in crosses with Flemish than the Polish do. Skull-width is sufficiently involved in the general increase of all bodily dimensions to more than offset the specific tendency of Polish to transmit a broad skull; for in skull-length (table 6), the Himalayan-Flemish cross-breds exceed the Polish-Flemish cross-breds even more than in skull-width. The amount of variability in skull-width, as indicated by the standard deviation (tables 7 and 8), is too erratic to have any par- ticular significance. F, is not uniformly more variable than F! in these cases. GENETIC STUDIES OF RABBITS AND RATS. 17 In studying bone-measurements, the length of 3 leg-bones has been investigated, that of the tibia, the femur, and the humerus (tables 9 to 11). The bones measured are from the right side of the body. The results are very similar in all three cases. The Polish-Himalayan cross is followed by such increase in vigor that the cross-breds of both the F! and the F 2 generations surpass in bone-dimensions the intermediate between the parent breeds, but the Flemish crosses produce offspring which even in FI fall slightly below the interme- diate between the parent breeds, and in F 2 fall below it still more. In every case F 2 is more variable than FI. CORRELATION. The question may properly be raised whether the same genetic factors affecting size operate in all parts of the body, or whether there are special factors affecting the size of each part. The latter view is favored by Davenport in his studies of human stature; the former seemed to me to be indicated in the statistics of rabbit meas- urements published by MacDowell (1914). Wright (1918), from a statistical analysis of the same data, concludes that both general and special factors are indicated. The present investigation should be able to throw further light on the subject. If two parts or dimen- sions of the body are influenced by the same genetic agencies, they should vary in unison, as the genetic agencies are made to vary by means of cross-breeding. If they are influenced by different genetic agencies, they should vary independently of each other. For this reason it is desirable to study the correlation existing between each pair of size-characters studied. The results of such a study are contained in the correlation tables (tables 12 to 29). These show strong correlation in every case between size of the body as a whole and size of each of its parts, as well as between different parts of the body. Most of the correlation coefficients (table 30) lie between 0.80 and 0.90, where 1.00 would indicate complete identity of all agencies affecting size, whether genetic or non-genetic. The highest correlations are found in com- paring the lengths of the long bones of the legs, humerus with femur (0.906) and with tibia (0.904), and femur with tibia (0.927). Next in closeness of correlation comes the relation between skull-length and the length of the leg-bones (0.806 to 0.871), but the correlations of weight with bone-measurements are almost as close (0.820 to 0.852), except in the case of the tibia, where the coefficient falls to 0.758. There is a very similar range in the correlation coefficients between ear-lengths and weight (0.836) and ear-length and the various bone-measurements (0.823 to 0.836), except again in relation to the femur, where occurs the lowest correlation of the entire 15 studied, viz, 0.741. 18 GENETIC STUDIES OF RABBITS AND RATS. In the case of MacDowelTs observations (Castle, 1914), correlation coefficients were obtained somewhat lower than those here recorded, but still notably high. The leg-bone correlation coefficients (the highest of any in both sets of observations) were in MacDowell's rabbits 0.858, 0.857, and 0.791, where the present observations give 0.927, 0.906, and 0.904. The skull-length correlation with tibia was, in the former case, 0.701, in the latter 0.806. In general, the correlations run about 10 per cent higher in the present set of obser- vations, which is probably due to the greater range of variation in size in the lot of rabbits OP which these observations have been made. The number of rabbits studied in the two cases is comparable, a maximum of 376 in the case of MacDowell's rabbits, of 348 in the present case. But the range of size-differences is greater in the present case. For example, the tibia classes in MacDowell's rabbits range from 88 to 110 millimeters; the range in the present lot is from 80 to 112 millimeters. The correlation is strongest where the range is most extended, for at the ends of the range, where only individuals of pure race occur, genetic differences are greatest and non-genetic agencies sirk into insignificance, whereas at the middle of the range Don-genetic agencies exert a relatively greater influence. That this is so can be shown by a fuller analysis of one of the correlation tables. Let us take, for example, the correlation between femur and humerus (table 24). The correlation coefficient (r) for all the rabbits (343) taken collectively is 0.906. For the pure races only (table 25) it is 0.980, almost perfect correlation, indicating nothing but genetic agencies at work. For the Fj cross-breds (table 26), which should be no more variable than the more variable of the parent races, r is 0.888, much lower than for the pure races, because now only intermediate forms are present, the extremes represented by the pure parent races not being present. For F 2 by itself (table 27), r is 0.878, very nearly the same as for Fi, it will be observed, although the genetic diversity is greater. For F 2 is more variable than F! as regards both femur and humerus, yet the correlation indi- cated is practically the same, the difference being no greater than the probable error. If there were independent inheritance of factors affecting the size of femur and of humerus respectively, the correla- tion should be less close in F a than in FI because of recombination of independent genetic factors, but such is not the case. Hence we are forced to conclude that exactly the same genetic agencies affect the size of femur and humerus. Similar reasoning would lead to the conclusion that the same is true of all size features studied, in- cluding not only bone-dimensions, but also weight and ear-length. GENETIC STUDIES OF RABBITS AND RATS. 19 GENERAL OR LOCAL SIZE FACTORS. In the light of these facts, what should be our attitude toward the view (expressed by some students of human heredity) that mixed races are likely to contain individuals with physical maladjustments and disharmonies? On this view it is assumed that there are inde- pendent genetic determiners affecting the size of the different parts of the body. If so, when races of different size are crossed, recom- bination of genetic factors will in later generations produce some parts larger, others smaller, than the general average of the races crossed. This will result in disharmonic combinations, as, for example, hearts too large or alimentary tracts too short for the bodies in which they are found. Is there any real ground for such appre- hension? I think not. There is in health a perfect correlation in size of each part with every other part of the same body and with the size of the body as a whole. The modern sciences of embryology and physiology tell us why this is so. It is (1) because development of the individual from the fertilized egg is so largely epigenetic, each stage growing out of its immediate predecessor; and (2) because it is so largely controlled by internal secretions. The view of the genetic independence in size of the various parts of the body is a sporadic relapse into preformationism, such as was perhaps excusable to the Grecian mind when, without the control of observational or experimental science, it fancied animals to arise by chance coming together of arms, legs, and other parts which originally floated free and unconnected in primordial slime. The day for such preposterous ideas is past. There may be valid reasons why mixing of the more distinct human races should not be advocated, reasons perhaps sociological, but there need be no fear that an animal or- ganism will result whose parts are not properly coordinated. We are only beginning to understand the mechanism of such coordination, through studies of the ductless glands, but it is already clear that such coordination and control are very complete and are adequate for the production of harmonic organisms in the widest racial crossing possible in the animal kingdom. I think that a strong probability has been established that the genetic factors which affect size in mammals are general in their action, exclusively so. In this last particular I dissent from the view ex- pressed by Wright (1918) who, from a statistical examination of MacDowell's data, concluded that the genetic factors indicated were in minor part special and local in action. Davenport has advo- cated a similar view concerning the inheritance of human stature, but on grounds which appear to me to be inadequate for two reasons: first, because of the imperfect character of his data, and secondly because of unwarranted deductions from them. Measurements of the several elements of human stature made on the living subject 20 GENETIC STUDIES OF RABBITS AND RATS. are far from being precise and consequently correlations between these measurements are unreliable. Yet such is Davenport's entire material. He resolves the total stature (standing height) into four elements, only two of which are capable of direct measurement, viz, the "fibula" and the "torso." The first element of the total stature is the "fibula," which is the "height of fibula head (attachment of external lateral ligament) from the floor." The second element, the "femur," is obtained by subtracting from the standing height, first the "fibula," and then the "sitting height." The third element, the "torso," is obtained by a measurement from chair-bottom to upper end of sternum, and is likely to be vitiated (as is the sitting- height) by amount of flesh or fat or clothing on buttocks. The fourth element, the "head and neck," is arrived at by subtracting "torso" from "sitting height." There are altogether too many chances of error or inaccuracy in these measurements to make them comparable in reliability with measurements made on the actual skeletal elements of individuals. Davenport argues for the independent inheritance of the four elements of stature on the ground that races and families have characteristically different proportions of the total stature formed by each of its elements. In support of this view he figures (from "Martin, 1914") side by side photographs of a "Dinka negro" and a "Chiriguan Indian." The enlargement of these pictures is arranged so as to make the individuals of seemingly the same total height, whereby their difference in proportions is obvious. The limbs of the negro are seen to be relatively long, those of the Indian relatively short. If, now, each is a fair representative of his race, we must conclude that Dinka negroes inherit as a racial character limbs relatively long, while Chiriguan Indians inherit as a racial character limbs relatively short. But a comparison of the pictures shows also that the Indian stands much nearer the observer than the negro, as the size of his head, eyes, hands, and feet and the diameter of leg and arm are much greater. It is his nearness to the observer that makes him appear as tall as the negro in the picture. He is accord- ingly of absolutely short stature, whereas the negro is of absolutely tall stature. The absolutely tall individual (and race) accordingly has relatively long limbs, the absolutely short individual (and race) has relatively short limbs. Is this association accidental? If not, we may be dealing here with one racial difference, not two. Tall stature and relatively long limbs may be due to one and the same genetic cause, and not be independently inheritable. Daven- port's own paper indicates that this is probably true. He repro- duces another figure (fig. 8) from "Martin, 1914," in which it is shown by diagrams that the proportions of the human body change in the lifetime of the individual, the limbs becoming relatively longer GENETIC STUDIES OF RABBITS AND RATS. 21 as the absolute stature increases. There is the same difference in proportions between boy and man as between Chiriguan Indian and Dinka negro. The inheritance of boy and man is the same; one is a further development of the other. May it not be that tall and short human races are the result of interruptions at different stages of the general growth process? If so, races or families with " relatively long fibula" will be such simply because they are tall races or families, and both the tallness and the long fibula will be due to one and the same ontogenetic cause, long- continued growth. It is no accident, probably, but the result of a common genetic and ontogenetic agency, that South Italians are (1) short of stature, (2) short-limbed, and (3) mature (cease to grow) early, whereas Swedes and Scotch are (1) tall, (2) long-limbed, and (3) mature late. There is a similar difference between the Polish and Flemish breeds of rabbits. We have evidence of the most posi- tive sort that a single genetic agency is responsible in one case for (1) small size, (2) short ears, and (3) early maturity, and in the other for (1) large size, (2) long ears, and (3) late maturity. In one case we have initial energy of growth small (as evidenced by the form of growth-curve) and soon spent; in the other case, growth energy is strong and persistent. This single difference will account for all the closely correlated size-differences, respectively, of the small and of the large races of rabbits. What is the nature of this genetic agency? Is it a gene, or an assemblage of linked genes, or what is it? I do not think that we can give a full and final answer to these questions at present, but we can at least outline certain possibilities and exclude others. First, inheritance of large or small size in rabbits is influenced equally by the father and by the mother. No difference can be detected between the results of reciprocal matings between large- sized and small-sized races. This indicates that sperm no less than egg is the vehicle of transmission and makes it probable that the chromosomes are concerned in the transmission. If so, the agency may properly be a gene or genes. THE NUMBER OF SIZE GENES. It is clear from the results of crosses between large-sized and small- sized races of rabbits that more than one gene must be involved, since there is no reappearance in F 2 of the grand-parental size-classes. The question may be raised how many genes, supposing all to be of like influence on size, will account for the observed F 2 distribution. There are two ways in which one might attempt a statistical solution of this question. He might consider how frequently the grand- parental conditions (extremely large or extremely small) reappear in F 2 and make this a basis for estimating the number of independent 22 GENETIC STUDIES OF RABBITS AND RATS. genetic factors involved. Thus, reappearance of the extremely large type in 1 individual in 4 would indicate one genetic factor; its reappearance once in 16 individuals would indicate two factors, and so on; but in the present case the extremely large type has not reap- peared at all in F,, so that this method is scarcely applicable. On the multiple-factor theory, it may be that too small a number of F 2 individuals has been produced to make the reappearance of an extreme type probable; but the form of the variation curve for F 2 would still be available as an index of the number of factors involved, even if this curve was not sufficiently extended to include the grand -parental classes. With this idea in mind, I have suggested (Castle, 1921) a comparison of the standard deviation of F 2 with the standard deviation of Fi as a basis for estimating the number of independent factors involved in cases of so-called blending inheritance. Dr. Sewall Wright, who has kindly assisted in this matter, gives the fol- lowing formula for computing the number (n) of independent factors involved : In this formula D is the difference between the means of the parental (pure) races, a\ is the standard deviation of Fi, and <; GENETIC STUDIES OF RABBITS AND RATS. It can therefore be stated with confidence that no linkage is found between large or small size and the four coat-characters, albinism, dilution, yellow, and angora. A reason for this has already been suggested, that each of the coat-characters is determined by a single recessive gene borne probably in a single chromosome, and in the case of each of the four characters in a different chromosome, whereas size depends on many genes borne probably in many different chro- mosomes or in all the chromosomes. In table 33 is recorded the color of each rabbit studied, but an examination of these records reveals nothing essentially new as regards color inheritance. It does, however, serve to confirm the discovery made by Punnett (1912) of a peculiar form of the extension factor (E) which he observed in the offspring of a certain Himalayan rabbit. To be sure, Punnett did not call it a peculiar form of the extension factor, but rather a darkener (D) inseparably coupled with the extension factor (E), so that the supposed couplet, darkened extension (DE of Punnett), behaved as the allelomorph of ordinary extension (E) and of yellow (e). A series of 3 allelomorphs was thus established by Punnett's observations, and it would seem desirable for simplicity so to designate them. I have elsewhere employed the symbol E' for Punnett's darkened extension (DE). The order of dominance of the 3 allelomorphs is E' (dark extension), E (ordinary extension), e (restriction or yellow). Our observations on the Flemish crosses reveal the probable source of Punnett's peculiar darkened extension found in his Himala- yan doe 7. No Himalayans of pure race, that we have had, possessed the darkened extension, but it is regularly present in Flemish Giant rabbits of the varieties steel gray and black. Doubtless Punnett's Himalayan doe 7 was derived from a Flemish cross made with the idea of intensifying or darkening the pigmented markings of the Himalayan (nose, ears, tail, feet). All our pure Flemish rabbits have possessed darkened extension. The black doe, B, was apparently homozygous for this factor. She transmitted it to two young, d"2473 and 9 2474, which she had by the Polish male 3, whose formula was Ee. These two young are recorded as black, but as full-grown adults it was noted that each of them at times showed slight indications of ticking on the neck or front legs, and among their offspring were typical steel grays (e. g., 93108, 93111, c?3420), similar in appearance to 97 (plate 1, F). That animal was heterozygous for dark extension, her formula being E'E. Mated with the Polish buck 3, she had a litter of 9 young, of which 5 were gray, 2 steel-gray, and 2 black. The 5 gray never produced any steel-grays when mated inter se, which shows that they did not inherit darkened extension, but only ordinary extension, E. But the steel-gray and the black young produced GENETIC STUDIES OF RABBITS AND RATS. 27 steel-gray offspring in various matings, and this shows that they did inherit darkened extension, E r , from their mother. Their sire, as we have stated, did not transmit E f , but was of the formula Ee. The steel-gray Flemish buck 2, used in crosses with both Polish and Himalayan does, was also heterozygous for dark extension, his formula being E'E. By Himalayan does (EE) he had 8 steel-gray and 9 gray offspring (see table 33, VI). By Polish does (EE or Ee) he had 10 steel-gray and 6 gray young (table 33, V). Unlike 97, he had no black young. He was homozygous for the gray or agouti factor (A) whereas 97 was probably heterozygous for this factor. Hence the two black young of 9 7, her formula being E'EAa, but that of 6*2 being E'EAA. These results confirm the conclusions of Punnett and show that (1) Any individual which is homozygous for darkened extension is devoid of agouti ticking, whether or not the agouti factor is present. Examples are found in black Flemish and black Siberian rabbits. (2) Any individual which is heterozygous for dark extension (E'E or E'e) will ordinarily be steel gray in color ("agouti-black," Punnett) if the agouti factor is present, either heterozygous or homozygous, but such individuals are often black (showing no trace of agouti ticking) in their first coat, although they develop the ticking in later pelages. However, some individuals of formula E'EAa may show little or no ticking in their adult pelages, as, for example, the FI 9 2474 and her brother 6" 2473, which produced steel-gray young as well as black ones when mated with each other. If the allelomorph of E present in a heterozygous individual is e rather than E } the steel-gray is usually (perhaps always) clearly visible in the adult coat. Punnett speaks of individuals of this sort as invariably " black," but his classifications were apparently based exclusively on the juvenile coat. In the F 2 generation of the Flemish-Polish cross, but not of the Flemish-Himalayan cross, appeared dilute pigmented individuals having dark extension. These I have called ' ' steel-blue "and" blue. ' ' They correspond with the classes steel gray and black of the intense- pigmented series. From the fact that no dilutes appeared in F s from the Flemish-Himalayan cross, it is clear that dilution was introduced in the Polish race, but not in either of the other races. Similar reasoning shows that the color yellow (e) also was not present in either the Flemish or the Himalayan individuals employed in the crosses, but only in the Polish individuals; for yellow F 2 individuals appear in the Polish-Flemish cross, but not in the Himalayan-Flemish cross. Finally, angora coat appears in F 2 of the Polish-Flemish cross, but not in either of the other crosses. Even in the Polish-Flemish cross angora appears only in the F 2 of a single mating, that between Flemish 97 and Polish c?3. But c?3 was employed also in the Himalayan-Polish crosses, yet no angora 1?S GENETIC STUDIES OF RABBITS AND RATS. individuals appeared in those crosses. Therefore c?3 can not have transmitted angora coat. Accordingly, Flemish 97 must have done so. She was not employed in the Flemish-Himalayan cross, otherwise we should have expected to see angora individuals in the F, of that cross also. SUMMARY. 1. Studies have been made of the weight, ear-length, and several bone-dimensions of three races of rabbits, and of the first and second generation hybrids between these races. 2. Two of the races (Polish and Himalayan) are of small size, like the wild rabbit of Europe, the ancestral species. The third race (Flemish Giant) is very large, a racially new condition. 3. Crosses between the pure races produce in general individuals of intermediate size both in FI and F 2 , so that the inheritance is correctly described as blending. But when the parent races do not differ greatly in size (as, for example, the Polish and Himalayan races), the size of FI individuals may be increased by heterosis beyond what it would be through inheritance alone, so that the size of the larger pure race is approximated or even surpassed. Nevertheless, with the disappearance of the heterosis effect in F 2 the average size of the cross-breds sinks to a strictly intermediate position. 4. The heterosis effect is seen in the FI generation produced by crossing a large with a small race, no less than in the cross between two small races, but when the difference in size between the races crossed is large, the heterosis effect is not sufficient to obscure the essentially blending or intermediate character of the inheritance in FI. It merely produces a rise in the FI average size above the strictly intermediate position, which F 2 in every cross closely approaches. 5. The variability in size of F 2 is regularly greater than that of FI, which in accordance with the multiple-factor hypothesis is regarded as indicating the occurrence of genetic factors affecting size in several different chromosomes or linkage systems. 6. An attempt has been made by statistical methods to estimate how many different chromosomes (or linkage systems) are concerned in the inheritance of the size-differences found in the three race crosses, with the following average results: for the Polish-Himalayan cross, at least two chromosomes; for the Himalayan-Flemish cross, about eight chromosomes; for the Polish-Flemish cross, ten or more chromosomes. As the total number of chromosomes in the rabbit is estimated at 10 to 12 pairs, it seems probable that all chromosomes are concerned in size-inheritance. 7. A study has been made of the correlation between weight, ear- length, and the several bone-dimensions studied, with a view to dis- covering whether the same genetic agencies influence size in different GENETIC STUDIES OF RABBITS AND RATS. 29 parts of the body, or whether factors, independent one of another, govern size in different parts of the body. The conclusion is reached that the genetic agencies affecting size in rabbits are general in their action, influencing in the same general direction all parts of the body. The same would probably be found true for man and other mammals, perhaps for vertebrates in general, but not for plants in which hor- mone action is less in evidence. 8. No linkage relation has been found to exist between size and any simple (unifactorial) Mendelian character, such as albinism, yellow coat-color, dilution, or angora coat. This result is in har- mony with the view that unifactorial characters have their genes located each in a single chromosome, while size is influenced by genes located in many or all chromosomes. 9. In rabbits, size differences correlated with sex are very slight. In skeletal dimensions the adult male averages larger by 1 or 2 per cent, but in weight females surpass males, particularly in the larger breeds of rabbits; yet the differences are so small as not to disturb appreciably the correlation coefficients based on data in which both sexes are included. In ear-length no significant difference between the sexes can be detected. TABLES. TABLE 1. Average weight in grams of rabbits of the several groups studied, at various ages. Data used in the construction of figure 8. Group. No. Age in days. 30 308 605 363 377 350 40 414 422 870 459 478 521 60 549 574 1172 661 683 769 90 779 824 1800 1002 1210 1227 120 973 1049 2260 1307 1641 1802 150 1133 1240 2630 1580 1988 2154 180 1244 1423 2820 1752 2166 2386 210 240 1297 1659 1898 2388 2680 270 1311 1705 1927 2443 2719 300 1328 1753 1952 2466 2752 330 1344 1800 1962 2485 2789 360 1359 1806 3240 1971 2507 2826 Polish Himalayan Flemish Fi, PXH Fi. PXF 20 5 2 25 27 16 1284 1536 2940 1854 2283 2553 F,. PXH TABLE 2. Comparative weights in grams of the two sexes in the groups of rabbits studied. Group. No. of males. Av. wt. males. No. of females. Av. wt. females. Difference. Polish 10 1424 10 1403 - 21 Himalayan. 1 1880 5 1870 - 10 F,, PXH.. 15 1957 10 2005 + 48 F,, PXH.. 23 1560 27 1704 + 144 Fi, PXF... 14 2469 13 2545 + 76 F,, PXF... 61 2080 52 2176 + 96 F,, HXF.. 8 2694 8 2886 + 192 F,, HXF.. 30 2401 32 2531 + 130 30 GENETIC STUDIES OF RABBITS AND RATS. TABLE 3. Clarification at to weight of the several groups of rabbits studied. (Claw 10 includes weights 1.000 to 1,099 grams; class 11 includes weights 1,100 to 1,199, etc. There are no data for columns 11, 32, 33, 35 to 39, and they are therefore omitted.] Group. 10 1'2 13 14 15 10 1 1 a 10 2 17 IS 19 20 21 22 23 24 25 26 27 28 29 30 31 34 40 1 Average in grams. 1 OQ Polish Himalayan. Flemish (> 1 3 8 1 '> 3 1 7 1 1 1 20 6 3 25 50 27 112 16 62 1404 1866 3646 1973 1652 2512 2126 2827 2472 142 218 233 198 257 162 230 2 1 1 '2 Fi. PXH... Ft. PXH... F,, PXF... Ft PXF 1 i 7 13 4 7 9 6 4 10 i i 1C i 3 18 7 2 1 ] 1 1 11 4 5 12 7 5 13 15 4 4 10 5 1 2 3 2 F HXF 5 5 2 3 3 3 2 1 Ft HXF TABLE 4. Statistics of variation in millimeters of ear-length in the several groups of Pi, Fi, and F t rabbits studied. Race. No. Range. Average. Standard deviation. Intermediate between parent races. Polish 23 81 to 88 83 6 Himalayan 6 92 to 97 94 8 Flemish Pi, P X H 3 25 143 to 147 90 to 98 145.3 94 9 1 74 89 2 Ft, P X H 65 85 to 99 92 3 62 89 2 F,, H X F Ft.H X F 17 70 112 to 119 106 to 129 115.6 115 7 2.08 5 85 120.0 120 Fi, P X F 27 103 to 116 109 3 3 76 114 4 Ft, P X F 131 93 to 122 107 6 05 114 4 TABLE 5. Comparative average ear-length in millimeters of males and females in the groups of rabbits studied. Cross. Males. Females. Pi. P X H... 95.4 94.2 Ft, P XH.... 92.5 91.6 Pi, H X F. . . . 114.5 116.6 F,,H X F.... 115.7 116.5 F,. P X F.... 109.1 107.7 F,, P X F. ... 106.9 107.1 GENETIC STUDIES OF RABBITS AND RATS. 31 TABLE 6. Variation of skull-length in the several groups of rabbits studied. [All measurements in millimeters.] Race. No. Range. Average. Standard deviation. Intermediate between parent races. Polish 21 63 to 69 65 7 g 66 to 72 5 68 9 Flemish 5 82.5 to 88 85.5 F,, P XH 25 62 to 74. 5 70.2 2.16 67.3 F. P XH 53 65 to 75 68.9 2.25 67.3 F lf H X F 16 76 to 82. 5 78.9 1.79 77.2 Fi, H X F 64 69. 5 to 83. 5 76.5 2.85 77.2 Pi, P X F 27 73 to 79. 5 75.5 1.18 75.8 F,, P X F 125 63. 5 to 80 73.1 3.15 75.6 TABLE 7. Variation in millimeters of anterior skull-width in the several groups of rabbits studied. Race. No. Range. Average. Standard deviation. Polish 20 36 to 41 37 6 .25 Himalayan 8 35 to 39 5 37 07 Flemish 5 43 5 to 47 45 3 F,,P XH F,, P XH Fi.H X F F, H X F Fi, P X F Fj, P X F 25 53 17 64 27 125 36. 5 to 41. 5 35. 5 to 40. 5 39. 5 to 46 38. 5 to 45 40 to 44 37 to 45 39.4 38.2 42.9 41.6 42.3 40.8 .15 .35 .40 .36 .72 1.36 TABLE 8. Variation in millimeters of posterior skull-width in the several groups of rabbits studied. Race. No. Range. Average. Standard deviation. Polish Himalayan 20 8 36 to 40. 5 37 to 40 38.0 37.9 0.96 1.09 Flemish F lf P X H 5 25 44 to 47 36 5 to 42 5 45.4 39.5 1.12 .33 F 2 , P X H 52 35. 5 to 40 38.1 .17 F,,H X F Fi, H X F F,, P X F F,, P X F 17 64 27 125 40. 5 to 44. 5 38 to 45 39. 5 to 44 37 to 44. 5 1 43.2 42.0 42.2 40.8 .95 .36 .32 .20 32 GENETIC STUDIES OF RABBITS AND RATS. TABLE 9. Variation in millimeters of length of tibia in the several groups of rabbits studied. Race. No. Bang*. Average. Standard deviation. Intermediate between parent races. Polish 21 79 to 89 83.9 2.26 Himalayan 8 92 to 96 93.9 1.41 5 105 to 116 111.0 Fj. P X H 25 86 to 96 W.8 2.38 88.9 FI P X H 54 80 to 96 89.8 3 39 88 it Fi. H X F 17 95 to 105 99.0 2.44 102.4 FI. H X F 60 90 to 105 97.3 3.10 102.4 Fi. P X F 27 92 to 100 96.2 2.27 97.4 F f . P X F 127 82 to 104 93.4 4.75 97.4 TABLE 10. Variation in millimeters of length of femur in the several groups of rabbits studied. Race. No. Range. Average. Standard deviation. Intermediate between parent races. Poliah Himalayan . . 20 8 68 to 76 77 to 82 72.3 79 5 1.71 1 83 Flemish 5 94 to 102 97 6 Fi, P X H 25 73 to 82 78 2 1 92 75 9 Ft, P X H 54 68 to 80 76 2 2 49 75 9 Fj. H X F 17 84 to 91 87 2 1 61 88 5 Fj. H X F 6G 78 to 90 84 7 2 86 88 5 Fj, P X F 27 80 to 86 83 3 1 67 84 9 F,, P X F 126 71 to 88 80 4 3 83 84 9 TABLE 11. Variation in millimeters of length of humerus in the several groups of rabbits studied. Race. No. Range. Average. Standard deviation. Intermciliatc between parent races. Polish 21 54 5 to 60 5 57 7 1 39 Himalayan 8 61 5 to 63 5 63 75 Flemish. . 4 73 to 71 5 75 1.. P X H I. P X H 25 51 59 to 65 VI ."i to 64 5 62.8 60 9 1 51 2 37 60.3 60 3 FI. H X F 17 66 to 71 68 8 F. H X F F,. P X F Ft. P X F. f.6 27 126 62. 5 to 71. 5 63. 5 to 68. 5 56 to 70 6 67.3 66 2 no o 2.26 1.20 69.0 66.3 GENETIC STUDIES OF RABBITS AND RATS. 33 TABLE 12. Correlation of ear-length with weight. * * J3 ?. *.s a Ear-length in millimeters. 1 80 83 80 89 92 95 98 101 104 107 110 113 110 119 122 125 128 131 134 137 140 143 140 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Total . 2 1 3 1 3 2 3 3 1 2 3 1 i 3 1 2 3 1 1 1 1 2 4 4 4 6 3 1 1 1 1 2 1 4 4 (j 6 3 2 2 1 1 i 4 3 2 3 i 1 G 9 12 12 16 22 23 22 24 29 18 28 31 23 15 13 8 4 3 ::: 1 4 2 5 4 1 4 6 1 1 2 2 3 3 1 1 1 1 1 1 2 4 7 5 6 6 2 3 1 i 3 7 5 8 6 2 4 1 2 6 1 3 7 5 3 3 1 1 1 "3 3 3 2 3 2 4 2 1 1 1 2 5 4 1 1 1 1 i 1 i i 2 2 1 i i 1 3-2-2 ~ 1 6 13 8 14 20 r,i 16 29 14 38 38 33 24 16 5 3 3 2 Mean ear-length, 105.350.43. S. D. ear-length, 3.860.10. Mean weight, 2 14. 15 16.8. S. D. weight, 448.6 11. 9. r, 0.830 0.011. TABLE 13. Correlation of ear-length with skull-length. Skull- length. 80 Ear-length in millimeters. I 6 16 22 43 55 45 58 U 24 10 3 2 83 86 1 3 89 1 4 4 2 1 02 95 98 101 104 107 110 na 110 119 122 125 128 , 134 137 140 143 146 62 64 66 68 70 72 74 76 78 80 82 84 86 Total . 2 2 2 1 7 2 3 5 12 7 3 1 a 8 17 2 3 1 1 7 1 1 4 2 3 14 6 1 1 10 4 2 3 3 8 16 9 1 5 6 12 14 2 2 3 8 12 9 1 1 8 4 5 3 9 3 7 5 2 2 1 1 1 1 1 2 1 1 1 1 3 t ~~ 1 6 13 X 14 27 34 17 3(\ 17 40 41 82 n 17 A 3 2 7 Mean ear-length, 105.45 0.42. S. D. ear-length, 11. 46 0.28. Mean skull-length, 73.030.16. S. D. skull-length, 4.420.11. r, 0.8360.011. 34 GENETIC STUDIES OF RABBITS AND RATS. TABUS 14. Correlation of ear-length with humerus-length. H Ear-length. | SO 1 5 83 SO 89 1 1 4 92 95 98 101 104 107 110 113 116 119 122 125 128 131 134 i:57 140 143 146 64 60 68 00 02 04 00 08 70 72 74 70 Total. 1 4 1 2 ? 1 ; | 2 ? i i 4 17 23 41 78 67 58 32 16 1 1 1 j :: 3 3 4 3 7 14 3 21 7 4 5 3 6 13 9 6 3 7 2 4 12 10 11 2 1 1 6 19 12 3 7 14 8 7 4 9 8 2 1 8 6 3 1 3 j 3 2 1 "l 7 13 8 13 27 34 16 30 18 41 41 36 24 17 5 3 3 1 2 339 Mean ear-length, 105.46 0.41. S. D. ear-length, 11. 46 0.29. Mean humenu, 64. 1 7 0. 13. 8. D. humerus, 3.73 0.09. r, 0.823 0.01 1 . TABLE 15. Correlation of ear-length with femur-length. ?emur- length. Ear-length. 1 SO S3 hO S9 J2 95 98 101 104 107 110 113 11C 119 122 125 128 131 134 137 140 143 146 08 70 72 74 70 78 80 82 84 80 88 00 02 04 00 08 Total 2 3 2 1 1 9 1 1 1 4 6 24 23 48 48 48 55 31 30 21 2 1 343 2 5 1 3 1 7 1 6 7 10 2 2 2 9 11 7 2 3 1 8 2 1 | 1 1 | 7 I 1 2 2 4 I 6 2 2 4 11 8 9 5 4 1 3 4 10 11 7 6 4 10 6 11 4 \ 2 7 4 4 7 6 5 4 1 i i 2 2 2 1 1 2 i i 1 7 1 :w 17 30 41 2 13 8 14 28 17 43 36 24 17 6 3 3 1 Mean ear-length, 106.46 0.41. 8. D. ear-length, 11.47 0.29. Mean fcmur, 80.730.18. 8. D. femur, 6.000.12. r, 0.8280.011. GENETIC STUDIES OP RABBITS AND RATS. TABLE 16. Correlation of ear-length with tibia-length. 35 Tibia- length. Ear-length. 1 80 83 86 89 92 95 98 101 i 3 5 4 6 5 3 3 104 107 110 113 116 119 122 125 128 131 134 137 140 143 140 80 82 84 86 88 90 92 94 96 98 100 102 104 106 108 110 112 Total. 3 4 1 1 7 3 1 1 1 2 4 1 5 1 4 3 1 2 3 4 8 6 4 i i i 2 6 10 8 5 3 4 4 3 2 5 10 13 22 29 42 65 54 41 37 23 7 4 3 1 5 3 4 i 2 3 8 10 6 5 5 2 2 1 4 3 9 7 8 9 4 8 6 7 6 3 2 1 4 5 7 7 "5 5 2 4 1 "3 2 i 1 2 1 1 ... 1 1 1 2 7 13 8 15 27 34 17 30 17 43 41 36 24 17 6 3 3 1 1 344 Mean ear-length, 105.42 0.41. S. D. ear-length, 11. 48 0.29. Mean tibia, 93.64 0.19. S. D. tibia, 5.27 0.13. r,. 0.741 0.016. TABLE 17. Correlation of weight with skull-length. Skull- length. Weight in hektograms. Total. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 34 40 62 64 66 68 70 72 74 76 78 80 82 84 86 Total. ? 1 1 1 2 i 5 13 20 39 51 43 56 62 23 11 2 2 1 2 1 4 2 3 2 5 2 1 1 1 4 5 2 3 4 4 2 1 1 2 7 1 2 1 2 4 1 i i 2 i i i 3 9 8 2 's 8 5 1 1 5 10 4 10 6 4 3 i 6 6 12 2 3 30 ... 1 5 7 5 6 8 7 1 2 2 5 7 13 4 1 2 6 7 6 2 5 4 3 1 1 5 9 12 12 16 23 22 21 24 18 26 32 23 u 13 8 4 3 2 i 318 Mean weight, 2,141.016.9. S. D. weight, 447.011.9. Mean skull-length, 73.17 0.16. S. D. skull-length, 4.380.11. r, 0.8520.010. 36 GENETIC STUDIES OF RABBITS AND RATS. TABLE 18. Correlation of weight with humerus. Hum*. Weight. Total. rua. 10 11 1-2 13 14 i:> If, 17 IN 19 M 21 n 23 24 25 26 27 28 M M 34 40 64 66 68 60 62 64 66 68 70 72 74 76 Total. 1 o 3 16 21 39 72 59 58 31 16 1 1 1 i 3 I 1 1 2 2 1 4 4 1 3 1 '2 2 1 i 5 1 4 1 6 7 8 2 1 1 7 11 I 7 9 4 4 9 1 4 2 6 14 8 1 8 7 3 6 4 11 3 2 6 7 10 4 5 6 9 2 2 4 6 2 2 6 6 1 1 3 3 1 2 2 1 1 1 :;j 28 i I 7 13 11 I.', 23 2-2 I'D IB 30 18 27 n 14 13 s 4 3 2 1 318 Mean weight, 2,146.5 16.8. S. D. weight. 445.9. Mean humerus, 64.260.14. S. D. humerus, 3.750.10. r, 0.8200.012. TABLE 19. Correlation of weight with femur. Femur. Weight. Total. 10 11 13 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 34 40 68 70 72 74 76 78 80 82 84 86 88 00 02 04 06 08 Total. l 1 a 2 1 1 3 2 2 i 4 2 3 1 2 3 4 2 1 4 3 (i 2 4 5 21 22 45 44 43 52 28 29 20 2 3 9 Q 4 1 3 1 9 2 2 1 2 8 7 5 1 1 6 6 6 9 4 2 6 7 8 1 3 1 6 6 8 3 2 2 2 5 10 4 5 2 j 2 6 7 4 4 2 5 1 5 3 3 5 3 2 2 3 1 i 1 1 2 4 3 1 1 14 30 1-2 Hi 18 l (i 8 1-2 23 22 19 25 27 31 M 13 8 1 i 318 Mean weight, 2,143.016.9. 8. D. weight, 448.011.9. Mean femur, 80.800.19. S. D. femur, 5.030.13. r, 0.841 0.011. GENETIC STUDIES OF RABBITS AND RATS. 37 TABLE 20. Correlation of weight with tibia. Tibia. Weight. Total. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 34 40 80 82 84 86 88 90 92 94 96 98 100 102 104 112 Total . i 2 1 2 1 2 1 2 1 3 4 2 2 1 2 1 1 2 1 1 1 3 1 3 4 3 1 1 5 5 4 5 4 1 i i 2 7 7 4 4 3 ? 1 3 7 6 6 1 2 3 6 1 ? i 2 2 4 4 2 3 2 4 9 12 20 27 39 61 48 38 37 22 7 4 2 2 1 2 8 3 4 3 4 2 6 4 5 2 2 7 5 2 2 2 5 3 7 3 3 1 2 4 8 2 9 4 1 1 2 6 5 1 2 1 2 1 1 1 1 1 ,1 i 6 20 25 27 32 8 4 3 i 7 13 12 16 23 22 30 18 23 14 13 2 1 320 Mean weight, 2,143.616.8. S. D. weight, 446.611.9. Mean tibia, 93.77 0.20. S. D. tibia, 5.29 0.14. r, 0.758 0.015. TABLE 21. Correlation of humerus-length with skull-length. Humerus-length. Skull- length. Total. 54 56 58 60 62 64 66 68 70 72 74 76 62 2 1 5 64 1 5 5 5 16 66 4 4 7 3 7 25 68 5 I 13 14 2 42 70 14 23 14 3 54 72 5 22 11 8 1 47 74 1 2 9 23 18 3 2 58 76 3 12 21 11 5 52 78 2 6 12 4 24 80 4 4 2 10 82 1 2 1 4 84 1 1 2 86 1 1 Total . 24 78 64 60 31 1 1 5 16 42 16 2 340 Mean humerus, 64. 16 0.13. S. D. humerus, 3.77 0.09. Mean skull, 72.960.16. S. D. skull, 4.460.11. r, 0.834 0.011. 38 GENETIC STUDIES OF RABBITS AND RATS. TABLE 22. Correlation of femur-length with skull-length. Skull- length. Femur-length. Total. 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 62 64 66 68 70 72 74 76 78 80 82 84 86 88 Total. 1 1 2 1 1 4 3 9 7 5 16 25 43 54 47 58 53 24 10 4 2 1 1 4 4 11 2 1 1 8 12 15 10 2 i 9 16 12 6 2 4 15 9 16 4 1 6 10 18 20 1 3 9 10 6 2 1 4 14 9 i 2 I 6 7 2 1 1 1 1 i 1 1 ... 4 6 25 23 48 46 49 55 30 29 21 2 2 1 1 i 343 Mean femur, 80.80 0.19. 8. D. femur, 5.20 0.13. Mean skull, 73.060.16. S. D. skull, 4.530.11. r, 0.871 0.008. TABLE VS. Correlation of tibia-length with skull-length. Skull- length. Tibia-length. Total. 80 82 84 86 88 90 92 94 96 98 100 102 104 106 112 116 62 64 66 68 70 72 74 76 78 80 84 86 88 Total. 1 1 2 2 2 4 3 1 7 4 1 1 1 3 5 5 2 5 2 2 13 2 4 3 1 .... 2 13 13 8 4 2 'e 4 17 13 9 5 1 5 16 25 43 54 46 58 53 24 10 14 2 1 4 13 5 18 13 5 5 10 13 7 1 1 5 7 12 7 3 i 7 4 5 5 2 4 1 1 2 1 j . 1 4 11 14 21 27 43 55 53 41 35 23 7 4 1 2 1 342 Mean tibia, 93.740.20. S. D. tibia, 5.450.14. Mean kulJ. 73.060.16. 8. D. skull, 4.540.11. r, 0.8060.015. GENETIC STUDIES OF RABBITS AND RATS. TABLE 24. Correlation of femur with humerus, all group* included. 39 Hume- rus. Femur. Total. 68 70 72 74 76 78 80 82 84 86 88. 90 92 94 96 98 54 56 58 60 62 64 66 68 70 72 74 76 Total . 3 1 1 5 4 18 24 41 79 64 61 32 16 2 1 10 13 2 2 10 11 'is' 28 2 V 29 8 2 7 14 6 1 11 8 1 18 26 4 4 25 26 i 23 7 1 1 2 2 25 4 7 23 48 46 49 55 31 29 20 2 2 1 343 Mean femur, 80.69 0.18. S. D. femur, 5.08dh0.13. Mean humerus, 64. 17 0.13. S. D. humorus, 3.78 0.09. r, 0.906 0.006. TABLE 25. Correlation of femur with humerus, pure races only. Hume- rus. Femur. Total. 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 54 56 58 60 62 64 66 68 70 72 74 76 Total . 1 1 1 4 1 6 4 1 2 11 6 3 7 1 > 2 1 3 1 2 :: i i 2 1 1 1 2 6 11 1 4 1 3 2 1 i 33 Mean femur, 76.95. S. D. femur, 8.06. Mean humenis, 60.95. S. D. humerus, 5.60. r, 0.980 0.004. 40 GENETIC STUDIES OF RABBITS AND RATS. TABLE 26. Correlation of femur with humerus, F, only. Hume- rus. Femur. Total. 72 74 76 78 80 82 84 86 88 90 58 60 62 64 66 68 70 Total. 1 e' 6 1 9 1 1 4 17 14 19 11 3 2 6 1 8 2 3 5 1 3 6 4 2 1 .... .... 1 2 7 11 12 11 9 6 1 N Mean femur, 82.64. S. D. femur, 4.08. Mean humerus, 65.58. S. D. humeruB, 2.68. r, 0.888 0.017. TABLE 27. Correlation of femur with humerus, F t only. Hume- rus. Femur. Total. 68 70 72 74 76 78 80 82 84 86 88 90 64 66 68 60 62 64 66 68 70 Total. 2 2 7 17 34 55 50 42 21 13 4 8 2 9 9 16 20 2 8 20 6 12 21 3 3 19 20 14 6 2 4 8 6 1 7 6 1 2 1 13 20 37 34 37 42 20 20 14 1 241 Mean femur, 80.67. S. D. femur. 4.48. Mean humenifl. 66.21. S. D. humerus, 3.45. r, 0.878 0.010. TABLE 28. Correlation of tibia with humerus. Hume- I'ibia Total rus. 80 82 84 86 88 90 92 94 96 98 100 102 104 1M 112 54... ? 3 5 56 3 7 4 2 1 17 58 1 7 12 4 24 60 3 5 14 17 3 42 62 3 7 19 29 15 4 2 79 64 . . 2 5 19 20 13 6 65 66 4 15 21 12 g 60 68 3 3 12 10 4 32 70 4 5 3 3 16 72 j 1 2 74 1 76 1 . Total. 6 11 14 22 28 41 66 53 42 36 23 7 4 1 2 344 Mean tibia, 93.64 0.19. 8. D. tibia. 5.35 0.13. Mean humerus. 64.150.13. S. D. humerus. 3.78 () >, 0.9040.006. GENETIC STUDIES OF RABBITS AND RATS. TABLE 29. Correlation of tibia with femur. 41 1 Tibia. 1 80 82 84 86 88 90 92 94 96 98 100 102 104 106 112 116 68... 70... 72 2 3 1 3 7 2 11 7 3 8 25 23 49 48 49 55 31 29 21 2 74.. . 1 10 5 10 12 5 1 2 18 18 5 76... 78... 80... 82... 84... 86 14 17 18 6 6 17 25 5 2 7 18 11 2 1 ..... 6 9 17 4 6 7 10 3 3 1 2 1 88... 90... 92... 94... 96... 98... 102 1 1 1 2 1 1 1 1 1 1 Total. 5 11 14 22 28 43 56 53 41 37 23 7 4 1 2 1 348 Mean tibia, 93.69 0.19. S. D. tibia, 5.48 0.14. Mean femur, 80.76 0.18. S. D. femur, 5. 19 0.12. r, 0.927 0.005. TABLE 30. Correlation coefficients derived from tables 12 to 9. Table. Characters correlated. Coefficient. Table. Characters correlated. Coefficient. 12 Ear-length, weight . . . 0.8360.011 22 Femur, skull-length. . 0.871 0.008 13 Ear-length, skull 23 Tibia, skull-length . . . 0.806 0.015 length 0.8360.011 24 Femur, humerus 0.906 0.006 14 Ear-length, humerus . 0.823 0.011 25 Femur, humerus(pure 15 Ear-length, femur. . . . 0.828 0.011 races only) 0.980 0.004 16 Ear-length, tibia 0.741 0.016 26 Femur, humerus (Fi 17 Weight, skull-length. . 0.852 0.010 only) 0.888 0.017 18 Weight, humerus .... 0.820 0.012 27 Femur, humerus (Fi 19 Weight, femur 0.8410.011 28 only) 0.878 0.010 20 Weight, tibia 0.758 0.015 28 Tibia, humerus 0.904 0.006 21 Humerus, skull-length 0.834 0.011 29 Tibia, femur 0.927 0.005 TABLE 31. Number of size factors indicated by each set of measurements as differentiating the three races crossed. Race. Tibia. Femur. Humerus. Skull- length. Weight. Ear-length. Av. Polish-Himalayan . . 2.1 2.6 1.0 3.1 1.6 1.5 2.0 Himalayan-Flemish. 9.8 7.3 4.8 7.0 9.7 10.6 8.2 Polish-Flemish 4.7 5.2 5.3 5.6 21.0 21.2 10.5 |J GENETIC STUDIES OF RABBITS AND RATS. TABLE 32. Comparative ize of the two sexes in the several groups of cross-bred rabbits. Group. Tibia. Femur. Humerus. Skull-length. Males. Females. Males. Females. Males. Females. Males. Females. t i c | 6 1 1 a i 6 y. 1 a y. i 1 c y. 1 Fi PXH 15 93.65 24 90.07 8100.20 33 97.78 14 96.59 65 94.29 10 :50 9 88 18 1,2 91.65 89.68 97.90 96.88 95.75 92 1 1 15 24 a 5:5 11 M 78.78 76.45 87.95 84.90 83.52 80.87 10 80 9 88 18 62 77.45 75.97 86.68 84.63 83.15 79.53 15 22 8 53 14 05 63.53 61.15 69.45 67.36 66.41 64.03 10 29 9 5:5 13 01 62.20 60.73 68.26 67.33 65.93 63.88 15 23 52 11 15 70.50 69.18 79.48 76.37 75.59 73.14 10 30 9 152 13 80 69.95 68.80 78.58 76.66 75.50 73.12 F,.PXH FI. HXF Ft. H XF F,. PXF Ft. PXF Weighted mean*. Per cent greater. 94.72 1.58 93.24 81.35 1.10 80.46 64.66 0.57 64.29 73.42 0.16 73.30 Anterior skull-width. Posterior skull-width. Ear-length. Weight. Group. Males. Females. Males. Females. Males. Females. Males. Females. I 1 1 c 1 4 1 d 1 d y. I 6 z 1 | 1 F,. PXH ] F^PXH S Fi.HXF F,, H XF i F,. PXF 1 F,, PXF ( 5159 90 38. 65 843.63 1141.91 442.48 (441.28 10 .'51 9 83 13 01 38.65 37.94 42.36 41.38 42.13 40.32 15 2.'5 8 51 11 64 40.23 38.61 43.88 42.49 42.84 41.20 10 29 1 33 18 01 38.35 37.90 42.70 41.50 41.58 40.38 15 24 8 57 14 07 9 9 11 11 10 10 54 .25 .45 50 .91 6 10 31 9 '53 13 04 9.42 9.16 11.66 11.65 10.97 10.71 15 28 x 50 14 il 1, 1. 2, 2. 2, 2, 957 500 094 401 '.! (WO K 27 8 32 18 52 to to to to -> to 005 704 886 531 545 176 Weighted mean* Per cent greater. 41.05 2.03 40.23 41 2 21 40 40.22 10 55 10 .5H .20 2, 110 2, 221 5.26 The weighted means were obtained by counting each male mean the same number of times as each female mean, the number of individuals assumed for the purpose of the computation to occur in each sex of each gioup being as follows: FI, PXH, 15; F, PXH, 30; Fi, H XF, 10; F,, HXF. 33; F,, PXF. 15; F,. PXF. 66. GENETIC STUDIES OF RABBITS AND RATS. 43 TABLE 33. Descriptive list of rabbits studied. Abbreviations in column headed "color": A, snow-white albino; An, angora; B, black; Dil, dilute; G, gray; H, Himalayan albino; St. G, steel gray; Y, yellow. Measurements in milli- meters; weight in grams. I. POLISH, ALL SNOW-WHITE ALBINOS. Designation. Mother. Father. Born. -3 MI 3 ja CQ Anterior skull- width. Posterior skull- width. 3 1 | W 1 1 91 92 99 910 92413 92712 92713 92738 9 2753 93186 1 10 10 2 9 2738 2738 2753 2753 2753 3 3 3 3 3 3 3 3 3 3 June 11, 1917 Sept. 10, 1917 Do. Sept. 13, 1917 Sept. 20, 1917 June 14, 1918 Do. Aug. 16, 1918 Do. Do. 68.9 68.4 68.2 65.5 65.4 66.2 63 .6 63.6 69.3 65.0 65*6 67.0 63.4 65.5 67.1 64.5 64.6 37.3 38.6 38.4 37.8 36.2 36.5 36.5 37.5 41.0 39.6 39.2 37.9 38.7 38.2 36.8 38.9 38.8 37.8 38.5 38.4 38.8 37.5 37.8 37*8 37.4 40.5 38.4 38^9 37.7 38.7 38.3 37.1 37.3 39.6 75.4 73.1 73.4 73.9 68.4 70.5 71.0 71.5 69.9 76.4 73.2 72.5 73.0 72.9 71.0 73.3 73.6 72.8 8e'4 84.7 86.0 84.9 80.1 80.7 81.4 84.5 79.9 89.6 84.2 83.6 87.6 82.7 82.3 83.6 84.5 85.0 59^3 57.1 58.8 60.3 55.5 57.0 57.3 58.2 56.7 60.9 57.5 57^2 58.8 57.7 57.7 57.6 57.5 58.8 8.4 8.3 8.4 8.7 8.3 8.1 8.7 8.1 8.5 8.2 8.5 8.4 8.4 8.4 8.4 8.2 8.2 8.1 8.4 8.4 8.2 1.275 1,425 1,545 1,575 1,250 1.355 1,295 1^230 1.280 1.700 1.450 1^505 1,400 1.450 1.450 1,200 1,340 1.325 93187 93488 9 3489 93490 o*3 0*2312 0*2313 0*2314 o"2412 0*2741 0*2742 0*2751 0*3318 o"3487 0*3561 10 10 10 1 2 2 9 2713 2753 2712 3 3 3 3 3 3 3 3 3 3 May 28, 1917 Do. Do. June 11, 1917 Sept. 13, 1917 Do. Sept. 20, 1917 June 17, 1918 Aug. 16, 1918 Aug. 21, 1918 II. HIMALAYAN, ALL "HIMALAYAN" ALBINOS. i "5 .M ]j Designation. 1 | Born. I l| ll s S 1 I > 1 1 i < * * B = i 94 66 1 35 8 37 3 77.3 92.9 61 .". 96 72.9 39.6 40.2 K2.S 95.9 63.9 9.7 2.210 92206 4 5 Apr. 27, 1917 69.1]37.3 37.4 79.3 94.3 62.4 9.2 2,085 92210 4 5 Do. 69.0137.3 38.0 80.7 94.8 63.8 9.5 1,835 9 3565 2210 2208 Aug. 21, 1918 67.5|35.9 37.4 77.8 92.1 63.9 9.5 1,620 93566 2210 2208 Do. 66.9135.3 . 72 1 37 3 37.1 37 4 77.0 81.3 92.9 96.0 62.8 ft? 9 9.6 1,600 c?2208 4 5 Apr. 27.1917 69.138.2 39.2 80.0 93.0163.6 9.4 1.850 1 ] | 44 GENETIC STUDIES OF RABBITS AND RATS. TABLE 33. Descriptive list of rabbits studied Continued. III. FLIMIBH GIANT. Designa- tion. , | j Born. 1 Anterior skull- width. Posterior skull- width. 1 H 1 1 9B 97 cfA <72 c?2681 B StG StG StG StG 82.9 87.1 88.2 84.6 85.6 44.5 45.4 46.7 43.8 47.4 44.7 44.0 47.4 44.9 46.4 95.3 97.8 102.6 94.1 98.5 106.7 113.1 116.4 105.2 113.5 73.1 75.6 73^4 77.9 14.7 14'e 14.3 4,060 3^480 3,400 'Sept."8.'l917 9B rf2 IV. Fi, POLISH X HIMALAYAN, ALL "HIMALAYAN" ALBINOS. Designation. Mother. 1 1 Born. Skull-length Anterior skull- width. Posterior skull- width. j .2 .0 W 1 1 92317 92320 9P 9P 4H 2P 2P 6H 10P 2206 H 2210 H 2210H 9P 9P 9P 4H 4H 4H 2P 2P 2P 6H 6H 6H 6H 6H 2210H 5H 5H 3P 5H 5H 3P 5H 3P 3P 3P 5H 5H 5H 3P 3P 3P 5H 5H 5H 3P 3P 3P 3P 3P 3P May 28, 1917 Do. Do. May 29, 1917 Do. June 4,1917 June 28, 1917 Apr. 27, 1918 Do. Do. May 28, 1917 Do. Do. May 28, 1917 Do. Do. May 29, 1917 Do. Do. June 4, 1917 Do. Do. Do. Do. Apr. 27, 1918 74.8 72.3 70.5 70.5 70.8 68.7 70.7 62.1 67.8 71.4 74.5 70.4 72.7 69.0 71.0 71.7 68.1 68.5 70 9 68.5 71.3 70.6 70.8 70.7 68.4 39.9 39.6 38.7 38.8 39.8 38.5 38.7 37.0 36.9 38.8 41.0 39.0 40.2 39.4 40.7 40.9 39.4 38.9 40 4 41.8 40.9 39.3 39.9 40.6 38.4 38.6 38.5 38.5 39.4 39.6 38.2 36.8 37.3 37.5 38.7 40.2 38.8 40.0 39.6 40.9 40.6 40.2 37.8 40 5 42.7 41.4 39.4 40.7 41.0 39.9 79.0 78.6 79.0 76.9 77.8 78.5 78.4 73.9 75.4 77.9 79.9 78.4 78.0 77.9 79.7 82.0 76.5 75.5 77.3 78.9 81.9 78.7 80.5 80.7 77.4 95.4 93.2 93.7 91.7 92.7 91.9 91.5 86.0 89.2 91.8 96.5 92.9 93.4 93.9 96.8 96.8 91.2 90.3 92.7 91.7 96.8 92.9 94.0 95.5 91.6 64.7 63.3 63.7 62.2 63.0 62.5 61.6 59.3 60.7 60.9 64.6 63.0 62.8 62.6 63.7 65.1 62.7 61.9 62.4 62.4 64.5 62.6 63.7 63 8 62.8 9.6 9.4 9.6 9.5 9.7 9.5 9.2 9.4 9.3 9.7 9.5 9.3 9.5 9.5 9.7 9.4 9.4 98 9.4 9.7 9.6 9.5 9.5 9.6 2,270 2,285 1,940 2,390 2.495 1,940 1.675 l!540 1,825 2,145 1.945 1,770 1.835 1,920 2.060 1,945 1.945 2,110 2,175 2,045 1.865 1.870 2,000 1.735 9 2323 92340 . 92341 92395 92446 93094 93099 93100 c?2316 d"2318 d"2319 d"2321 d"2322 d"2324 c?2342 d"2343 d"2344 c?2391 tf2392 cf2393 d"2394 rf2396 d*3101 GENETIC STUDIES OF RABBITS AND RATS. 45 TABLE 33. Descriptive list of rabbits studied Continued. V. Fi, POLISH X FLEMISH. Designa- tion. Color. 1 1 Born. C3 -2 "3 QQ terior skull- dth" 3193 StG 2430 2434 June 14, 1918 72.8 40.0 40. 82.5 95. 64. 10. 1,880 0"3194 B 2430 2434 Do. 70.9 40. 39.9 78.5 88. 61. 10. 2.020 c73196 Blue 2430 2434 Do. 75.0 41.8 40.6 83. 98. 65. 10. 2.175 d*3197 G 2576 2506 Do. 74. 42.5 42.0 80. 94. 63. 10. 2.250 (73199 B 2576 2506 Do. 74. 42. 42. 78. 90. 62. 10. 2,080 (73200 A 2576 2506 Do. 73. 41. 40. 78. 89. 62. 10. 1,930 C73222 B An 2432 2434 Do. 78. 41. 42. 84. 100. 66. 11. 2,160 d*3239 G 2451 2506 June 15, 1918 72. 42. 42. 77. 87. 61. 10. 1,895 0^3241 StG 2451 2506 Do. 77. 43. 42. 81. 94. 65. 11. 2,200 <73298 A 2436 2434 June 16, 1918 76. 41. 42. 88. 104. 70. 11. 2.410 cf3299 A An 2436 2434 Do. 72. 41. 10. 80. 95. 65. 10. 1,790 0*3300 A 2436 2434 Do. 72. 41. 41. 82. 98. 64. 10. 2.260 (73302 StG 2436 2434 Do. 75. 41. 41. 81. 93. 64. 10. 2,260 <73304 StG 2510 2506 June 17, 1918 79. 45. 44. 85. 97. 69. 12. 2.515 (73367 StG 2437 2434 July 18, 1918 73. 41. 40. 83. 99. 63. 11. 2,280 C73368 G 2437 2434 Do. 75. 42. 43. 84. 100. 67. 10. 2,390 0*3396 A 2510 2506 July 23, 1918 70. 40 40. 82. 95. 65. 10. c73398 StG 2510 2506 Do. 72. 42. 43. 79. 93. 64. 11. 2,125 d*3391 B 2510 2506 Do. 69. 39. 41. 79. 92. 62. 10. c73400 StG 2510 2506 Do. 72. 40. 41. 77. 86. 62. 10. 1,880 <73420 StG 2474 2473 July 4, 1918 69. 39. 39. 76. 89. 61. 9. 1,760 C73448 StG 2451 2506 Do. 77. 42. 42. 86. 97. 70. 12. 2,450 0*3449 G 2451 2506 Do. 74. 42. 41. 80. 92. 65. 10. 2,180 <73480 G 2576 2506 Do. 74. 41. 41. 79. 92. 61. 10. 2.125 (7348 B 2576 2506 Do. 74. 42. 41. 78. 90. 61. 10. 1,975 cf3529 StG 2511 2506 Do. 71. 40. 40. 80. 92. 64. 10. 2.130 o"3533 StG 2511 2506 Do. 73. 40. 41. 82. 95. 66. 11. 2.300 0*3534 B 2511 2506 Do. 71. 40. 41. 80. 96. 67. 10. 2,195 GENETIC STUDIES OF RABBITS AND RATS. 49 TABLE 33. Descriptive lift of rabbits studied Continued. VIII. F 2 , POLISH X FLEMISH Continued. Designa- tion. Color. Mother. Father. Born. Skull-length. Anterior skull-width. 1 Posterior skull-width. J 1 a 1 i 1 ^3603 A An 2430 2434 Aug. 1, 1918 75.0 42.7 41.6 88.3 101.9 69.0 11.7 2.335 d"3604 B 2430 2434 Do. 76.3 42.2 41.7 87.5 100.5 69.1 11.4 2.485 c?3656 A An 2436 2434 Nov. 2, 1918 74.6 41.2 41.4 83.8 96.2 64.6 10.2 2,110 cT3660 StG 2436 2434 Do. 71.1 39.6 40.7 80.9 93.5 64.2 9.6 2.150 d"3662 A 2432 2434 Do. 79.9 41.5 42.4 87.3 101.2 69.7 12.1 2,450 d"3664 Blue 2432 2434 Do. 71.7 40.7 41.3 82.2 96.8 65.5 9.9 1,900 c?3665 B An 2432 2434 Do. 77.8 42.2 41.9 86.9 102.2 68.4 10.9 2.450 d"3838 StG 2511 2866 Feb. 25, 1919 67.9 39.3 40.8 79.5 93.5 62.7 10.3 1,920 cT3841 G 2451 2866 Feb. 27, 1919 75.2 41.0 41.8 75.7 88.2 58.8 10.8 1.645 c?3849 A 2576 2866 Feb. 25, 1919 68.9 40.0 41.0 74.9 88.4 59.6 9.8 1.730 d"3851 B 2576 2866 Do. 70.7 41.5 40.1 78.0 91.1 61.8 10.4 2.025 cf3853 BlueG 2576 2866 Do. 71.1 41.4 41.2 78.7 90.6 61.2 10.7 1.870 c?3859 B 2510 2866 Do. 71.6 39.9 41.0 79.0 90.7 60.5 10.3 1.850 d"3878 A 2437 2434 Feb. 24, 1919 72.0 41.5 41.8 81.9 98.1 63.5 10.3 2.000 cf3882 G 2509 2866 Feb. 25, 1919 69.3 40.8 41.5 78.9 90.9 64.3 11.2 1,970 C"3889 DilY 2436 2434 Feb. 24, 1919 73.5 42.5 42.8 86.0 99.5 65.4 10.7 2,100 d"4035 StG 2511 2866 May 5, 1919 73.4 39.6 40.4 78.1 93.6 63.4 10.8 2,010 c?4037 A 2511 2866 Do. 67.6 38.2 37.8 71.3 84.9 56.3 9.3 1,675 c?4066 StG 2436 2433 Do. 70.3 41.7 40.8 80.3 95.4 62.8 9.8 1.800 c?4068 YAn 2432 2434 Mar. 3, 1919 73.0 40.8 40.9 85.2 101.2 66.3 11.5 1.920 IX. Fj, HIMALA YAN X FLEMISH. ' _L 1 | Designa- tion. Color. . Born. 1 L J^. , 1 1 3 a! || 3 Q ;| H 1? I t & 3* I 5 & & 1 92983 H 2646 2647 Apr. 18, 1918 74.6 41.8 41.5 81.9 93.9 64.4 10.8 2,220 92985 G 2646 2647 Do. 78.8 38.9 40.4 83.7 96.4 65.3 12.3 2,125 92987 B 2646 2647 Do. 80.8 40.7 41.0 88.3 101.4 67.6 11.9 2,535 92988 B 2646 2647 Do. 76.5 41.8 41.8 82.4 94.7 64.0 11.3 2,400 93024 B 2517 2522 Apr. 20, 1918 75.0 41.8 40.4 84.7 98.2 68.4 11.6 2.675 93025 G 2517 2522 Do. 76.9 41.8 42.0 82.9 96.3 66.1 11.7 2,900 93026 StG 2517 2522 Do. 73.0 41.3 42.8 84.4 96.7 67.5 12.2 2.590 93054 G 2521 2522 Apr. 22, 1918 78.5 40.1 39.7 84.4 97.9 66.6 11.3 2.160 93149 G 2517 2522 June 14. 1918 75.9 41.5 40.7 82.9 94.0 66.8 12.1 2.680 93150 StG 2517 2522 Do. 76.0 40.8 41.7 84.4 94.9 68.0 12.0 2,480 93151 StG 2517 2522 Do. 76.1 42.6 41.8 87.3 98.9 69.7 12.7 2.380 93153 H 2521 2522 Do. 77.1 41.4 40.5 88.8 101.3 71.4 12.1 2.705 93157 B 2521 2522 Do. 74.4 41.5 41.5 83.9 97.3 67.4 11.7 2.290 93202 B 2646 2647 Do. 78.3 41.9 42.7 81.9 93.2 64.0 11.1 2.440 93228 StG 2645 2647 Do. 77.5 44.7 43.4 85.4 97.0 69.4 12.0 2,880 93373 G 2519 2522 July 20, 1918 75.4 39.6 39.6 81.9 95.7 65.4 11.7 2.400 93375 G 2519 2522 Do. 77.2 40.2 40.8 82.6 93.0 65.4 11.3 2.070 93385 G 2642 2647 June 27. 1918 77.3 40.5 40.5 82.4 95.9 65.5 11.0 2.850 93387 StG 2642 2647 Do. 10. S 93389 StG 2642 2647 Do. 74.4 40.9 41L3 81A 95.0 ee's 11.3 2^520 93439 H 2517 2522 Aug. 16, 1918 76.8 42.1 42.2 87.3 99.6 69.6 11.1 2.750 93441 StG 2517 2522 Do. 69.8 38.9 38.4 79.2 90.9 63.1 10 9 2.260 93442 StG 2517 2522 Do. 75.9 41.9 40.4 87.9 100.5 71.0 11.8 2.900 50 GENETIC STUDIES OF RABBITS AND RATS. TABLE 33. Descriptive list of rabbits studied Continued. IX. Ft, HIM ALATANX FLEMISH Continued. DwiKn.i- tion. Color. | 1 Born. Skull-length. Anterior skull- 1 width. Posterior skull- 1 width. Femur. 08 3 H Humerus. a W J3 . 93443 B 2517 2522 Aug. 16, 1918 79.1 42.3 42.9 87.5 101.0 69.8 12.6 2,530 93444 8tG 2517 2522 Do. 73.3 40.9 41.9 82.7 93.8 67.6 11.1 2,410 93505 G 2645 2647 Do. 76.1 38.7 40.7 84.0 95.6 67.3 11.0 2.440 93523 G 2521 2522 Do. 78.0 43.3 41.7 86.5 99.1 69.6 11.5 2,560 93526 G 2521 2522 Do. 74.0 40.1 40.8 81.5 92.5 65.0 11.0 2,275 93570 G 2830 2647 Aug. 21, 1918 83.8 45.3 45.2 89.0 100.5 69.2 11 .8 2.710 93596 G 2646 2647 Sept. 1, 1918 77.8 42.8 42.7 87.3 98.6 67.3 12.8 2.750 93600 H 2646 2647 Do. 81.3 43.5 43.8 89.5 100.0 69.7 12.7 2,990 93601 G 2646 2647 Do. 79.8 42.7 42.9 88.7 101.7 70.8 11.8 2,620 93647 H 2519 2522 Nov. 2, 1918 74.2 41.0 42.4 83.5 95.5 64.6 11.6 2,500 (73023 H 2517 2522 Apr. 20, 1918 12.1 C73056 G 2521 2522 Apr. 22, 1918 76!s 41.2 4l!i 86.1 99.5 12.2 (73057 G 2521 2522 Do. 79.3 41.7 41.9 85.1 97.4 66^3 11.3 (73145 H 2517 2522 June 14, 1918 74.8 41.0 41.6 84.6 96.2 67.3 11.2 2,185 (73147 StG 2517 2522 Do. 74.8 42.2 42.5 88.4 95.4 67.5 11.9 2,170 rf-3148 StG 2517 2522 Do. 72.5 41.9 42.4 82.6 96.4 67.7 11.3 2,190 (73156 G 2521 2522 Do. 73.8 40.9 41.2 83.2 96.5 67.1 10.8 2.300 C73158 B 2521 2522 Do. 74.3 41.3 41.2 81.2 94.2 64.9 11.1 2,300 C73204 G 2646 2647 Do. 76.6 41.8 42.5 83.9 97.4 66.2 12.0 2,455 (73206 G 2646 2647 Do. 73.6 42.8 43.1 78.9 93.0 63.1 10.7 2,450 (73208 G 2646 2647 Do. 73.1 81.5 96.6 64.6 11.1 <73227 StG 2645 2647 Do. 76.9 43^4 44.0 83.7 96.0 66.4 11.5 2,480 <73372 H 2519 2522 July 20, 1918 84.9 98.3 67.7 11.6 2,350 c73386 StG 2642 2647 June 27, 1918 10.8 C73388 StG 2642 2647 Do. 10.8 (73445 StG 2517 2522 Aug. 16, 1918 75^7 41.9 42^6 87.2 106 'o 70'2 11.3 2,470 C73446 B 2517 2522 Do. 78.0 41.8 41.5 90.3 102.8 71.0 12.1 2,440 c73496 G 2831 2647 Do. 76.2 43.4 43.0 86.7 98.1 70.7 11.3 2,555 <73497 G 2831 2647 Do. 77.0 42.3 42.7 86.9 99.9 70.8 10.9 2,690 (73503 H 2645 2647 Do. 77.9 42.1 42.5 82.9 93.2 64.6 11.8 2,460 (73504 H 2645 2647 Do. 77.8 43.2 44.0 83.4 95.8 66.4 11.0 2,490 (73506 G 2645 2647 Do. 76.2 40.6 42.4 80.3 91.5 62.7 11.2 2,430 c73507 G 2645 2647 Do. 77.6 40.6 42.6 85.4 95.5 68.2 12.0 2,390 (73508 G 2645 2647 Do. 81.2 44.2 43.1 88.8 100.7 69.8 12.8 2.740 (73509 StG 2645 2647 Do. 70.8 40.4 41.3 80.8 90.3 64.9 11.2 2,190 <73520 B 2521 2522 Do. 70.4 40.9 40.0 81.9 95.6 65.0 10.6 2.070 (73524 G 2521 2522 Do. 76.1 40.4 40.3 86.0 99.8 67.2 11.2 2.170 (73525 G 2521 2522 Do. 76.3 41.3 41.6 86.1 98.9 67.8 11.4 2.305 (73568 G 2830 2647 Aug. 21, 1918 80.3 42.3 42.7 85.2 98.1 67.2 11.1 2.420 (f8M9 G 2830 2647 Do. 64.0 10.6 C73697 G 2646 2647 Sept. 1, 1918 8CK3 41.8 42.6 88.6 101.6 68.1 11.6 2.525 4*8606 G 2646 2647 Do. 79.9 42.0 43.2 84.6 98.7 67.5 12.0 2.500 (P85M G 2646 2647 Do. 80.3 41.9 44.7 89.5 101.9 71.2 12.0 2,350 c73602 G 2646 2647 Do. 83.7 44.0 45.0 89.7 105.4 71.5 12.4 2,870 rf'SMI H 2519 2522 Nov. 2, 1918 73.0 43.4 43.0 88.0 99.8 70.3 12.3 2,595 i?M4i H 2519 2522 Do. 77.2 43.9 44.4 89.4 103.2 69.8 12.9 2.640 <73650 G 2519 2522 Do. 71.9 40.7 40.9 82.7 99.1 66.2 11.6 2.070 PART II, ON A NON-TRANSMISSIBLE TRI-COLOR VARIATION IN RATS. In the course of studies of linkage in rats, I have observed the production of a tri-color individual, gray, yellow, and white, similar to the tri-color varieties of guinea-pig, and I had hoped to establish from it a new color variety of rat, but thus far no success has at- tended my efforts. The case, in its bearings on the nature and origin of variations, is not without interest, and so will be described briefly. The two genes whose linkage relations were being investigated when the tri-color individual made its appearance have been desig- nated c and p. Both are recessive in crosses and thus become visible as somatic characters only when present in the homozygous state, cc or pp respectively. An individual of the formula cc is an albino; an individual of the formula pp is pink-eyed and yellow-coated. When an ordinary albino is crossed with a pink-eyed yellow indi- vidual, young are produced which are neither albinos nor pink-eyed yellow, since neither c nor p will be homozygous in the cross-bred individuals, which are in fact gray in color like wild rats, or else gray-hooded, if the gene for hooded pattern is present in homozygous condition. By means of such crosses between albino and pink-eyed yellow rats, gray and gray-hooded young were being produced when the tri-color individual made its appearance. It is a gray-hooded individual, like its brothers and sisters, except that the areas nor- mally gray are liberally mottled with yellow. The mottling extends practically throughout the colored portions of the coat from nose to tail tip. The yellow areas vary in size from those which contain merely a few yellow hairs to those of a square inch in extent. The least yellow is found on the right shoulder, which is almost like normal gray in appearance. The most yellow is found in the middle of the back, to the left of the median line, where a large spot of clear yellow occurs in the wide back-stripe of the hooded pattern, which would correspond roughly with grade +2j/ of the grading scale used in our studies of hooded rats. The tri-color individual is, fortunately for the purposes of genetic study, a male. He has been mated with albinos, with pink-eyed yellows, and with gray FI cross-breds between the albino and the pink-eyed yellow races, and later with his own daughters of the colors white, pink-eyed yellow, and gray. Hundreds of young have been produced by these matings, for the animal is a remarkably large and vigorous one and his mates have proved very prolific, but 51 52 GENETIC STUDIES OF RABBITS AND RATS. not one of the young was mottled like the sire. This occasioned no surprise when the mates were not related to the tri-color male or were not his direct descendants, for it was conjectured that he might be a mutant due to a new recessive gene, which accordingly would only become visible when in a homozygous state. But if he him- self represented such a homozygous recessive combination, all his gametes should transmit the mottled condition, and all his daugh- ters consequently should be heterozygous for the mottled condition, and in matings with their sire should produce 50 per cent of mottled individuals. The fact that they do not produce mottled individuals shows this hypothesis to be untenable. The tri-color male, in fact, breeds like his gray-hooded brothers and sisters, producing gametes which are commonly either c or p, but which in no case transmit a mosaic relationship. From his ancestry and from the results of his matings, we know him to be a double heterozygote, CcPp. The Mendelian expectation is that he will form four kinds of gametes, cP, Cp, CP, and cp, but since there exists linkage (in this case repulsion) between c and p, the first two classes of gametes will be commoner than the other two. His breeding behavior accords with these expectations. In most of his gametes he transmits either albinism or pink-eyed yellow. In an occasional gamete he probably transmits both, a matter which has not been tested, as it would require special matings. It is certain that in an occasional gamete he transmits neither c nor p, a condition expressed hi the formula CP, which like cp would represent a cross- over. His gametes, accordingly, appear to be such as are normally produced by Fi doubly heterozygous individuals like his gray and gray-hooded brothers and sisters. How, then, can we account for the production of yellow areas in the gray coat? By an explanation similar to that which Morgan and Bridges (1919 1 ) have given to account for the production of gynandromorphs in Drosophila. We may suppose that at an early cleavage (perhaps the first) of the fertilized egg from which this rat developed, one of the pair of chromosomes, in which the genes c and p are borne, failed to divide as normally, or that for some other reason it failed to pass into one of the cell-products. Further con- sideration shows that it must have been the maternal rather than the paternal chromosome of this pair which was lost. For the mother was an albino and must have contributed cP, but the father was a pink-eyed yellow, Cp, and his contribution by itself would produce yellow fur, whereas with the mother's contribution it would pro- duce gray. We may suppose, accordingly, that a blastomere con- taining only the paternal contribution produced that part of the > Carnegie Inat. Wash. Pub. No. 278. GENETIC STUDIES OF RABBITS AND RATS. 53 skin which is yellow, and that a blastomere or blastomeres contain- ing as normally both maternal and paternal members of this chromo- some pair produced the gray part of the coat. But how about the germ-cells of this individual? Does the mixed condition seen in the coat obtain also among the germ-cells? Very likely it does. But if so, no tri-color progeny need be expected in consequence. For those germ-cells which correspond in composition with the gray parts of the coat would produce gametes like those of any other FI individual, which for the most part would transmit either albinism or pink-eyed yellow, with an occasional gamete transmitting gray and an occasional gamete transmitting both albin- ism and pink-eyed yellow. But the germ-cells corresponding to the yellow parts of the coat would transmit either pink-eyed yellow or nothing (so far as the chromosome under discussion is concerned). It is an open question whether this last type of gamete (lacking the entire chromosome) would be viable, and if it were, its existence would be difficult to detect, since it would behave like the double recessive type of gamete, cp, in its effects on coat color. How, then, could a genuine genetically transmissible type of mottled rat arise? Only, I suppose, by a change in the genetic locus at P, p, so that it became, instead of P alone or p alone, a mosaic Pp transmissible as a unit and capable of functioning as the allelomorph either of P or of p. Such apparently is the case in the guinea-pig, where Ibsen has shown the mottled condition to be the allelomorph of both simple black and simple yellow. Such appar- ently is also the condition in the mottled variety of rabbit called "Japanese," which I have recently studied in a series of crosses, and such would appear to be the case in maize with striped pericarp, which has been extensively studied by Emerson and his associates. They find that the mosaic factor frequently becomes simple by mutation either to all red or to all white, and that mutation occurs oftenest to that component of the mosaic which is most extensive in the parent. This result is easily understood if we suppose that the mosaic "gene" divides in such a way as to include more of one of its components than of the other in a cell product, or so as to contain only the major component in a cell product. That cell would then develop into a self-colored individual. I think that this idea of a mosaic gene will explain the high degree of variability which is found not only in striped maize, but also in spotted mam- mals, whether spotted with yellow or with white. Hooded rats and Dutch rabbits are examples of spotted mammals which I have stud- ied extensively and have found to be amenable to selection like the striped pericarp of maize studied by Emerson and by Hays, probably for the same reason, viz, because a mosaic gene varies (or, if you prefer, "mutates") in the proportions of its constituents. 54 GENETIC STUDIES OF RABBITS AND RATS. In 1912 1 I described a peculiar guinea-pig, which, in the light of the present case, I am inclined to explain in the same way as the tri-color rat, as a non-transmissible though not of necessity a purely somatic mutation. At that time I characterized the guinea-pig in question as a "pink-eyed individual with a colored coat," but this is not a very good description for it. The description fits better a genuine genetically transmissible variety that was then unknown to us, but which curiously enough was soon to appear in our breeding- pens from stock that had recently been obtained in Peru. The indi- vidual, whose peculiarities were not transmitted to its offspring, was by pedigree a heterozygote between albinism and pale black. Such a heterozygote I believe it in fact to have been, but in appearance it was white, except for some small spots of pale black ("blue") on the right side of the head and on the hips. The head pigmentation extended as a "faint pigmented streak" on to "the iris of each eye." The mother of this animal was blue, the father cream (which is recessive to blue), but he also transmitted albinism. He can not have transmitted blue, since that is dominant to cream. Undoubt- edly, then, the egg from which this individual developed transmitted color (C) and blue (E) which lie in different chromosomes. But the father transmitted albinism (c) and cream (e), which are reces- sive allelomorphs of C and E respectively. Evidently it was the maternal chromosome bearing C which was lost from the greater part of the ectoderm of the embryo, since no other loss would have produced an uncolored coat and eye. Whether or not E was lost in the same regions can not be determined, but since all colored areas were blue, not cream, it is evident that wherever C was re- tained, E was retained also. The peculiar individual was a female and produced three young, all albinos as I recall, by an albino sire, before her untimely death. It can not be stated, therefore, whether the germ-cells were hetero- zygous in nature or were like the greater part of the coat, pure albino, inherited from the sire alone. Certainly none of the three develop- ing ova transmitted the mosaic condition found in the mother. It seems probable, though it is regrettably unverifiable, that no young mottled like the mother would have been obtained, had we been able to obtain a much larger number of young. 1 Science. 36, p. 508. BIBLIOGRAPHY. CASTLE, W. E. 1914. The nature of size factors as indicated by a study of correlation. Carnegie Inst. Wash. Pub. No. 196, pp. 51-55. . 1921. On a method of estimating the number of genetic factors concerned in cases of blending inheritance. Science, 54, pp. 93-96, 223. in collaboration with H. E. WALTER, R. C. MULLENIX, and S. COBB. 1909. Studies of inheritance in rabbits. Carnegie Inst. Wash. Pub. No. 114, 70 pp., 4 pi. DAVENPORT, C. B. 1917. Inheritance of stature. Genetics, 2, 313-389. GALTON, F. 1889. Natural inheritance. 259 pp. HARVEY, ETHEL B. 1920. A review of the chromosome numbers in the metazoa II. Jour. Morph., 34, 1-68. HOSHINO, Y. 1915. On the inheritance of the flowering time in peas and rice. Jour. Col. Agr., Tohoku Imp. Univ., 6. LANG, A. 1910. Die Erblichkeitsverhaltnisse der Ohrenlange der Kaninchen, etc. Zeit. f. ind. Abst. u. Vererbungslehre, 4, 1-23. MACDOWELL, E. C. 1914. Size inheritance in rabbits. Carnegie Inst. Wash. Pub. No. 196, pp. 1-49, 9 figs. MILLER, G. S. 1912. Catalogue of the mammals of Western Europe. 1019 pp. British Museum, London. PUNNETT, R. C. 1912. Inheritance of coat-colour in rabbits. Jour. Genetics, 2, pp. 221-238, pi. 12-14. and P. G. BAILEY. 1914. On inheritance of weight in poultry. Jour. Genetics, 4, 23-39, pi. IV. -. 1915. Further experiments on the inheritance of coat-colour in rabbits. Jour. Genetics, 5, pp. 37-50. . 1918. Genetic studies in rabbits. I. On the inheritance of weight. Jour. Genetics 8, 1-25. WRIGHT, SEWALL. 1918. On the nature of size factors Genetics, 3, 367-374. 55 PLATE 1 P, Polish 9~2, mother of F, yonm: //, Himalayan 91, mother of FI yi /', Flemish (lianl V 7, color steel u The relative si/.e is ronerllv >ho\\: oosefl with hotli the uther races. in en. r- \vitli Imth the other races mother of FI young bv a Polish sire. :cent in the case of the Polish in- PLATE 2 Skull and certain leg Ixines of representative individuals of the three pure races of i.-il.liiis and of their Fj hybrids, shown on the same scale of reduction in size. To the left of the skull of each rabbit is seen the tibia-fibula, and to the right of the skull the femur of the ri^ht hind- leg. Below the skull is seen the right humerus. P, bones of Polish 9 9. F, bones of Flemish 97, shown alive in plate 1, /'. //, Uu,- ..i Himalayan 96. F u P X F, l>ones of 92436, daughter of Flemish 97 and of Polish o"3. F,, P X H, bones of 92320, daughter of Polish 99 and of Himalayan c?5. FI, H X F. Inn,.- ..f 92431, daughter of Himalayan 90 and Flemish cT2. 431 000583146 6 THE LIBRARY UNIVERSITY Of ( U IH)RM A Santa Barl>.n.i THIS BOOK IS DUE ON Till-. I \s I DATE 1