72 AZAEA'S DOG. 



be depended on for characterizing species, when, as in this case, not 

 accompanied by some weighty differences as to the skulls and teeth. 

 We therefore cannot feel justified in regarding this form as more than 

 a somewhat strongly coloured local variety of C. azarae. 



Another variety from the shores of the Straits of Magellan has been 

 distinguished by Philippi by the designation of C. patagonicus. He 

 makes its distinctness depend on its having a paler, thicker coat, with 

 a shorter tail marked with black and white rings. But it is in no 

 way surprising that an individual from so cold a region should have a 

 thicker, paler coat. We have found that specimens of C. azara from 

 the south have longer hair than specimens from Chile. As to the 

 caudal annulations, we have found two such to exist towards the root 

 of the tail of a specimen of C. azarcs from Patagonia. 



Philippi's species depends only on a single skin, without any skull. 

 It is therefore impossible for us to treat it as more than a local variety. 



Two species have been described by Lund under the names vetulus 

 and fulvicaudus, but they appear to us to be very imperfectly 

 distinguished. The former, he tells us, has the body and limbs 

 " slender ;" the latter has them " somewhat slender." The former, 

 is said to be above "light ashy grey;" the latter, "whitish 

 grey." The former is said to have the limbs below " isabel yellow ; " 

 the latter " brownish yellow." The only marked distinction asserted 

 is in the tail. That of vetulus is said to be black at the end and 

 for one fourth of its length; that of ftdvicattdits is described as 

 having the end and a patch upon its dorsum yellowish, and it is 

 also said that a patch of ochre-yellow exists behind the ear. This 

 distinction does not appear to us to be a satisfactory one, owing 

 to the variations we have found as to the colour of the tail in other 

 species. Burmeister has also remarked *, as a result of his experience, 

 that a ruddy tail-end forms no distinctive character, and he affirms 

 that Lund's two species approximate together strongly. This is 

 also our opinion. The fourth upper premolar of his C. fulvicaitdus 

 is small, but the dentition figured, has an aberrant appearance. 

 But Burmeister identifies Lund's C. vetulus with Wied's C. azarce ; 



* Fauna Brasiliens, p. 41. 



