PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 91 



of tlie complete iutergradation of the forms which they designate; it 

 will also indicate the number and names of the more i)ronounced races^ 

 I have been able to make out, arranged in chronological sequence, the 

 full syuonymy of each being given separately further on. 



There are uow before me 44 specimens of Scoi)Sowls from Tropi(;al 

 America, different specihcally from A', harharus, IS. Jiammeohai, nud ^. 

 nudipes^ and also very distinct from the hairy-toed members of the 

 genus. This series a|)pears, at hrst sight, to be n)ade up ot several 

 distinct species, there lieing no less than sis very i)ronounced types 

 of coloration represented ; these different styles being so exceedingly 

 different in appearance that in the absence of intermediate specimens 

 no one would hesitate to recognize their specific distinctness. 



These different styles are more or less characteristic of separate geo- 

 graphical areas; tlius, the '■^brasiliamts''^ type prevads over Eastern 

 South America, "MstM.s" in Upi)er Amazonia and in Columbia, '■'■(juate- 

 make''' iu Central America, and '•'cassini'^ in Eastern Mexico. They 

 thus partake somewhat of the nature of geographical races; were they 

 strictly such, the case would be very much simplified; but such, unfor- 

 tunately, is not the case, since it frequently occurs that extreme speci- 

 mens of one form may be found in a region of which it is not typical, 

 while several, if not all, of them may be represented in a sufihtiently 

 extensive series from a single district! Thus, we have true '• fpiatcmake^^ 

 from Bahia, Brazil; pure brasiliamis from Costa Eica and Guatemala; 

 and a specimen apparently very much like "wsiws" from Sta. Catarina, 

 S. E. Brazil. 



In the absence of specimens of neutral or intermediate character, 

 these facts would not be antagonistic to the theory of specific distinct- 

 ness of the forms named above, but, on the contrary, would be decidedly 

 confirmative, since they would do away with the probability that the 

 variations are purely the result of geographical impress. Generalized 

 specimens, however, or those which are not typical of either one or the 

 other of the several ri?ces, constitute a very large proportion of the 

 whole.* 



It is iu consideration of all these facts that we are led to conclude 

 that the several particular forms we have named above, however distinct 

 they nmy appear when the most specialized examples are compared, 

 are but " strains" of a single species, tending toward the establishment 

 of permanent geographical races (and in the course of time distinct spe- 

 cies), but which, in consequence of the non-extinction of specimens of 

 a generalized nature, have not yet passed the incipient stage. 



The variations in this species involvi^ not only differences in the 

 colors themselves, but in the character and distribution of the markings,, 

 scarcely two examples being exactly alike. Variations of a purely 



*A specimeu fiom Mazatlan, Western Mexico, in the gray phase, is so exactly inter- 

 mediate between guatcmahc and hraniUanus that it cannot be referred more properly to 

 one than to the other. 



