2 1 6 Bulletin American Museum of Natural History. [Vol. XXII , 



specimens (only 6 are measurable for this feature) from 20 to 23 mm., 

 averaging 21.4, and in the South American (n specimens) from 

 20 to 29 mm., averaging 27.8, with only one below 23 mm., while 

 only one of the Mexican skulls (the largest) reaches 23. This is 

 therefore a very constant and appreciable character, when it is con- 

 sidered t'hat it is nearly independent of the general size of the skull. 



While the Jaguar is unquestionably subject to a wide range of 

 individual variation in size and coloration (aside from the sporadic 

 black phase), as long since recorded by Wied for the Jaguar of southern 

 Brazil, it seems evident that were sufficient material available for 

 examination the Felis onca group would prove susceptible of separa- 

 tion into several easily recognized geographical forms; the material 

 for any such study is, however, at present lacking in the museums 

 of America. Dr. Mearns, as already stated, gave a number of charac- 

 ters which he believed, from an examination of the specimens in the 

 U. S. National Museum, would serve to distinguish the Mexican and 

 Central American Jaguars from those of South America. The alleged 

 difference in the width of the postpalatal fossa in examples from 

 north of the Isthmus of Panama as compared with South American 

 specimens is, as said above, fairly constant and of considerable im- 

 portance, but those based on the form of the bulla, on general size, 

 and on the relative size of the premolars prove to be too inconstant 

 to possess much value. His paper (cited above) on the Jaguars 

 had especial relation to those of Mexico and Central America, which 

 he considered specifically separable from those of South America, 

 and also susceptible of division into several species and subspecies. 

 Felis hernandezii (Gray) was considered as limited to the "arid 

 tropical areas of Mexico," while the name Felis hernandezii gold- 

 mani was proposed for the form inhabiting the "humid tropical 

 areas of Mexico," with the type locality, Yohatlan, Campeche. 

 This supposed form, however, does not seem separable from true 

 F. hernandezii (type locality, Mazatlan, Sinaloa), in view of the 

 wide range of variation shown by the Escuinapa series of true hernan- 

 dezii. His Central American Felis centralis, he calls the "smallest 

 of the Jaguars"; it is, however, quite as large as the specimens before 

 me from Venezuela and the Lower Amazon. In southern Brazil and 

 Paraguay the size is nearly the maximum, equalling, if not slightly 

 exceeding, the largest known specimens from Mexico. 



The subjoined table of skull measurements shows the great range 

 in size of Jaguars from Mexico, Costa Rica, Brazil, and southern 

 South America. 



