PHENOMENA AND PRINCIPLES OF BREEDING 471 



those specimens in whose progeny undesirable latent characters 

 would appear. Incidentally, the methods of studying breeding 

 problems which Mendelism has introduced are likely to lead to 

 important discoveries in relation to other phenomena of breeding. 

 Correlation of characters. If we have, to begin with, such an 

 individual as we desire, and the work is not obstructed by failure 

 of the individual of the desired type to breed, or by adverse pre- 

 potency of individuals mated with it, it is easy to fix or to eliminate 

 any single character, and this can be done in a very few genera- 

 tions ; but in breeding to fix, maintain, or produce a type, it is 

 necessary to consider many characters at the same time. If each 

 character, in its various expressions, were absolutely independent 

 of every other character, the making and maintaining of types ap- 

 proximating fixed standards would be a hopeless task. The char- 

 acters of an individual, being parts of an organism, are often 

 necessarily similar in certain manifestations, either throughout or in 

 closely related groups. The welfare of the individual depends to 

 a great extent upon the adaptation of its parts to each other and 

 to its conditions and mode of life. So there are established, in any 

 race or family bred on any principle of selection, certain apparent 

 correlations of parts occurring so regularly that, when considered 

 only where they occur, they appear to indicate an essential unity, 

 making the group of characters act as one. Thus, the body, legs, 

 neck, and head of a bird have as a rule a similarity (differing 

 outlines considered) of proportions ; a bird with long body is likely 

 to have a long neck, head, and legs ; a bird with very short, strong 

 bill and broad skull is likely to be short and heavily built throughout. 

 That these correlations are not essential is seen when we find in 

 such a variety as the Exhibition Game fowl an increase in length of 

 neck and legs quite out of proportion to the increase in length of 

 body, and in creeper varieties the size (including length) of body 

 maintained, while the length of neck is slightly, and the length of 

 legs greatly, reduced. Again there is a natural, general tendency to 

 correlation in structural character of bones, muscles, and skin. If 

 size and muscle are developed, making a large x heavy body, the 

 tendency is to coarseness throughout, coarse bone, coarse-fibered 

 flesh, and coarse, thick skin. But on examination of a number of 

 birds of this general type it will be found that there is not close 



