MURWM— SIGMODONTES— OCHETODON MEXICANUS. 129 



We have no doubt that Professor Sumichrast's specimens represent typi- 

 cally the Reithrodon mexicanus of Do Saussure. Though they present some 

 discrepancies in dimensions, De Saussure's measurements, lie says, are prob- 

 ably a little under the mark, while Sumichrast's specimens seem a little 

 stretched ; this is enough to bring about perfect concordance. 



Our Louisiana specimens are highly interesting, as showing for the first 

 time the occurrence of the true Mexican form in the United States. They 

 are unquestionably identical with Sumichrast's examples from Tehuacan, and 

 agree even better than these with De Saussure's figures. The discovery of 

 this style of Ochetodon in the United States is especially important in its bear- 

 ing upon the identification of Mus carolinensis, Aud. & Bach. Nos. 7748-7749 

 agree with the account of carolinensis in length of tail, and in the dullness of 

 color of the under parts ; and it may be, after all, that this long-tailed mexi- 

 canus ranges coastwise up to the Carolinas. Still, there are discrepancies 

 that cannot be overlooked between Audubon's description and the present 

 animal, especially as to the size of the feet ; and we are not at present war- 

 ranted in calling the form Ochetodon carolinensis. The latter name must be 

 assigned, with a query, as a synonym of humilis, at least until we find the 

 long-tailed large-footed form in Carolina. 



That there is a regular gradation in length of tail and size of foot 

 between humilis on the one hand and mexicana on the other is undeniable ; 

 and if we could have proven in the genus Ochetodon the same amount of varia- 

 tion with locality that has always been admitted in the case of Zapus hud- 

 sonius, and that obtains with Hesperomys leucopus and Arvicola riparius, we 

 should be forced to merge the three supposed species into one, with two 

 geographical variations. Failing in this, however, at present, we can, at any 

 rate, conveniently mark off three kinds of Ochetodon. The following analysis 

 (in which coloration, which though an aid in identification, may not be always 

 reliable, is omitted) will, it is believed, enable us to readily distinguish ninety 

 per cent, or more of our specimens : — 



A. Tail shorter than head and body (at most barely equaling head 

 and body). 



a. Hind feet under 0.70 long (usually 0.55-0.65) humilis. 



B. Tail longer than head and body. 



b. Hind feet under 0.70 (exceptionally just reaching 0.70) . longicauda. 



c. Hind feet over 0.70 (rarely, if ever, falling to 0.70). . - mexicanus. 

 9 M 



