MURIDJ3- SIGMODONTES— NEOTOMA. 1 3 



We tabulate none of the very young skulls before us, as these would 

 vitiate the results, especially as regards the lengthened muzzle of the bushy- 

 tailed species ; the young of that animal not having the snout noticeably 

 different from that of the rest. The difference in length of snout between 

 cinerea and jloridana does not seem to be much from the figures; but a tenth 

 of an inch on the end of a rat's nose is something; and, viewing the naked 

 skulls, the rostral part in cinerea is seen to be slenderer, as well as longer, 

 than it is in Jloridana ; the interorbital space is more constricted and more 

 deeply indented. Aside from this, the skulls show nothing diagnostic among 

 the several real or supposed species. Adult ones average about two inches 

 long (1.84 to 2.10) by an inch broad (0.84 to 1.12) across the zygomata, and 

 barely over two-thirds of an inch in height (taken as described — see under 

 Tab. 1, p. 12). The extreme length of the under jaw is about an inch and a 

 fourth on an average, but runs from 1.10 to 1.45; a part of this difference 

 being due to the fluctuating length of the under incisor. This generally 

 protrudes one-half of an inch, or a little more ; the upper incisors protrude 

 about one-third of an inch. The molar series is from three- to nearly four- 

 tenths of an inch long ; thereis nothing diagnostic in its length. 



In connection with the rostral elongation of the skull of JV. cinerea, we 

 should note another slight peculiarity — not, however, diagnostic of species, 

 much less of sections of the genus. In all the western skulls examined, includ- 

 ing even those of N. jloridana, there is a tendency to a narrowing of the con- 

 duit of the posterior nares by more or less filling in of bone from the alveolar 

 border. Thus, in ordinary South Atlantic Jloridana, the whole palate is defi- 

 cient behind the point indicated in the generic diagnosis — nothing bounds 

 the space on either side but the alve'olus itself. In the other extreme, there is 

 rpiite a little shelf on either hand, noticeably narrowing the aperture. But 

 the feature is extremely variable, and cannot be relied upon for even specific 

 diagnosis. 



We know three good North American species of this genus: first, there 

 is the ordinary N. Jloridana, really inhabiting most of the United States, 

 except New England, though decidedly southerly, and ranging thence into 

 Mexico ; secondly, the N. Juscipes, which appears to be perfectly distinct, 

 and is a highly interesting animal, opening the way, through several of its 

 features, into the tropical aspect of the genus, such as is exhibited by the 

 N. ferruginea of Tomes, from Guatemala, and furnishing an analogy to the 



