INTRODUCTORY. 5 



are graphically pictured in the colors of the thoroughly mongrelized domestic 

 pigeon, and even in some of the so-called "pure" breeds, described as "mottled," 

 "mealy," "grizzled," "splashed," "spangled," "well-broken," etc. 



It would seem to be obvious that 'purity of stock is the first essential for elimi- 

 nating "chance" phenomena. Purity is not to be found in domesticated forms, 

 nor yet in wild species that show intermingled forms. The terms "pure," "true," 

 "constant," as used by breeders, mean nothing more than a degree of uniformity 

 capable of being artificially maintained by careful selection and mating. The 

 best almond tumbler, for example, according to Fulton '" (p. 155) "will only remain 

 what it should be for at most two seasons." 



The purity requisite for scientific purposes is the self-sustaining ))urity found 

 only in unmixed natural species, such as are represented in large numbers and 

 convenient forms in the wild pigeons. Over 450 distinct species are recognized 

 in Salvadori's "Catalogue of Birds in the British Museum." In this large group, 

 undoubtedly arising from a single root and having much apparent homogeneity, 

 the species are remarkably distinct, even in nearly allied forms. Sometimes two 

 species are so nearly alike in size, form, and color that they appear, even to the 

 trained eye, to be one and the same. When the two are brought together in life, 

 fairly sharp distinctions are quickly detected in the voice, proportions, color, and 

 behavior, and crosses give hybrids that demonstrate specific distinctness. The 

 wild ring-dove of Japan and China (Streptojjelia torquatus)^ and the cage ring-dove 

 {Streptopelia risoria) are two such species. As St. risoria is not known in a ^ylld 

 state and is probably of African origin, and as St. torquatus enjoys geographic 

 isolation from it and its nearest relatives, it is safe to conclude that the two species 

 are pure with respect to each other, and their purity as regards other sources is 

 .not open to doubt. The hybrids from these two species are intermediates, but 

 incline somewhat towards torquatus, as is particularly manifest in voice and color. 



Although St. torquatus appears to be somewhat prepotent in crosses with St. 

 risoria, its superiority does not reach exclusive dominance in a single namable 

 character. Mendelian dominance does not exist, and the law that fails in the first 

 generation of course can not hold in the second. Dominance, so far as my obser- 

 vations go, is a thing of many degrees, and is far from representing a natural law. 

 Even allowing that it is a rigid rule in one or a few species, that would not raise 

 it to the rank of a law, if the test of a law is universal application. 



In crossing the wild passenger-pigeon (Ectopistes migratorius) with the cage 

 ring-dove (St. risoria), the hybrids have so far been invariably remarkably close 

 intermediates, and without exception sterile males. If this rule should hold in 

 experiments continued for a hundred years, it would still have no claim to being 

 a general law. At most it would only be a proved rule for a particular cross; in 

 another cross the blending may be only partial, both sexes may appear, and one 

 or both be fertile. The rule would be broken in every point, and obviously two 

 contradictory rules do not make one valid law. The regularity which we may 

 discover in individual cases, no matter with what detail, accuracy, and constancy 

 fulfilled, rises to the dignity of law only when referred to causes or processes of a 

 general nature. 



* Fulton's Book of Pigeons, London, 1S95. 



« Sirepiopelia douraca was later used as the proper name of the oriental rinR-dove. See Cluipter XM.— Editor. 



