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 ItaUans 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 Fj of the grand-parental size-classes. 

 The question may be raised how many genes, supposing all to be of 

 like influence on size, wiU account for the observed F2 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 

 Fi and make this a basis for estimating the number of independent 



