90 A TREATISE ON HORSE-BREEDING. 



oughbred. We have crossed in and out " with- 

 out rhyme or reason" until, with the single 

 exception of our thoroughbred horses, it is 

 scarcely possible to trace the pedigree of any 

 given American-bred horse four generations 

 back without finding an admixture of all the 

 various breeds and types that have ever been 

 known. With such an ancestry it is not to be 

 wondered at that disappointments meet the 

 novice on every hand. He selects a fine-look- 

 ing bay mare that will weigh nearly 1,500 Ibs., 

 in moderate flesh, clean-limbed and strong, and 

 he looks about for a stallion possessing the same 

 characteristics that he may couple the two to- 

 gether to produce a first-class draft horse. He 

 has been told that "like produces like" so often 

 that he believes it, and this theory leads him, 

 very properly, to think that from such a pair 

 his hopes of producing good draft horses will 

 be realized. But he is disappointed; the prod- 

 uce is not like either of the parents; and he 

 pronounces breeding a lottery, and the doctrine 

 of the transmission of the peculiarities of the 

 parents to the progeny a humbug. He forgets 

 that heredity transmits with certainty only 

 what has been firmly fixed in the ancestry; and 

 he loses sight of the fact that his large, fine 

 bay mare was herself the produce of a mixed 

 ancestry perhaps of a bay Clydesdale stallion 

 and a little sorrel mare of unknown blood, and 



