APPLICATION OF THE LAW OF HEREDITY. 9 



slvely Increasing numbers of ancestors In all earlier generations. From each of the 64 ancestors 

 in the fifth preceding generation an individual inherits only approximately one five-thousandtb 

 of Its characteristics. Hence the influence of this and earlier generations is so slight that it may 

 be disregarded. An undesirable feature which had not appeared in a stoclc for five genera- 

 tions is practically eliminated. 



The mathematical statement of this law has never — that I know of — been applied to poultry 

 breeding for the purpose of verification. Breeders familiar with the law observe that results 

 in breeding seem on the average to correspond very closely with the results indicated by this 

 law, and are satisfied to accept it. 



It should also be said that a parent does not transmit alike to all offspring. There is no abso- 

 lute rule ^ at least none that is known, and in breeding for special results expectations are 

 based on the probability that a certain proportion of the offspring of » mating will have the 

 qualities or characters sought. 



To illustrate:— Suppose a breeder of feather legged fowls finds in his flock a pullet that is 

 absolutely clean legged. We need not inquire here how such a pullet might be found in such a 

 flock, and be of the same breeding as the others. It might happen. From this pullet he con- 

 cludes to try to make a clean legged strain of the variety. He mates her with a male of the 

 same l)reeding having the lightest leg feathering. He hopes from such a mating to get a few 

 chickens bare legged like the hen. Most of the chickens he expects to come more or less 

 feathered on the legs. What would be of most use to him would be a cockerel clean legged 

 like hi" dam. 



Accord'ii^ to the general law stated above he may reasonably expect about one-fourth of the 

 progeny of this hen to inherit from her, but whether they are to be bare legged or inherit 

 other qualities is uncertain. But suppose he gets one bare legged cockerel. Next season he 

 mates this cockerel with bis dam, and may reasonably expect a good proportion of the progeny 

 of this mating to have clean legs, for both parents and one grandparent have that feature. 

 From this mating he may have clean legged specimens of both sexes, and mating these together 

 he may expect clean legs lo preponderate, for the parents, the grandparents, and half the great- 

 grandparents were clean legged. 



Now suppose that not seeking to make a clean legged stock the breeder carelessly, or to get 

 some other quality of the clean legged bird which he desires to fix in his stock, or, suppose that 

 by some chance mating of which he may not know the 151ood of a clean legged fowl is intro- 

 ■duced into a line of feather legged ones. Clean legs and scantily feathered legs are bound to 

 appear in the progeny for several generations, no matter how careful he may be in future mat- 

 ings, but if he avoids using them, and is careful not to introduce blood in which the same fault 

 appeared more recently than in the line from which he is trying to eliminate it, it will take him 

 only a few years to get it virtually stamped out. The rare chance of inherillug It from the 

 ancestor that brought it into the line may bring a clean legged specimen occasionally for many 

 generations, but they will not come in sufficient numbers to seriously alTect results. 



Let us make now the general application of the law we have examined as it affects a single 

 quality. 



The breeder of fowls has to deal with many desirable and many undesirable qualities. If he 

 lireeds systematically to definite standards for a number of years he secures a certain uniform- 

 ity of desirable qualities, and may also have a similar uniformity of undesirable qualities, 

 though from the fact that his selection with regard to undesirable qualities is a negative selec- 

 tion that is likely to be much greater variety in faults (as he considers them) than in merits. 



He may mate together a male and female that are as nearly alike as possible in every respect. 

 In proportion as they are bred on the same lines, and have the same ancestors or many of the 

 same ancestors in common, they may be expected to produce chicks uniformly like themselves. 

 But If they are entirely unrelated the chances of their reproducing their type are very much 

 reduced, and if — as is often the case — one or both of them came from stock in which the 

 ancestry presents a variety of different types, the results are apt to be very discouraging, for, 

 as the law given indicates, a fowl may hav* thirty difl'erent ancestors, each of which may have 

 an appreciable eflecton his inheritance of qualities. 



By inbreeding, by mating fowls bred on the same lines, the number of ancestors is reduced, 



