supraorbital border angular instead of rounded; anterior border of zygomatic plate convex in upper half and almost straight in lower half as opposed to nearly straight throughout in boylii; pterygoid fossa broader; bullae more pointed anteriorly and less inflated; mesostyles of upper molars larger; surface between orbital region and nasals convex in lateral view instead of flat.

Remarks.—When Saussure (1860:105) described P. aztecus he did not designate a type or type locality. Osgood (1909:157) designated as lectotype the mounted specimen, in the Geneva Museum, which has the skull inside and of which Saussure figured the molar teeth. Osgood (loc. cit.) examined one of the three specimens (No. 3926 USNM) that Saussure used in describing P. aztecus and found that it agreed "in every respect with recently collected specimens from Mirador, Veracruz, which, in the lack of exact knowledge, may be assumed to be the type locality, as it is certain that some at least of Saussure's specimens were taken near there."

Fig. 1. Two species of Peromyscus.
1. P. boylii ambiguus
2. P. boylii beatae
3. P. boylii levipes
4. P. aztecus (triangles)

Osgood regarded P. aztecus as a subspecies of P. boylii because of the resemblance between aztecus and P. b. evides, but evides is far removed geographically (occurring only in western México) from aztecus, and is smaller. P. aztecus is larger than any known subspecies of P. boylii, and is not known to intergrade with P. b. levipes or P. b. beatae (with which aztecus occurs sympatrically at Jalapa, Veracruz), the two subspecies of boylii that are found nearest the geographic range of P. aztecus. Also, as mentioned previously, aztecus possesses distinctive characters that distinguish it from all subspecies of boylii. For these reasons I regard aztecus as a distinct species.

According to Hall and Kelson (1959:634), P. aztecus occurs in San Luis Potosí, Hidalgo, and west-central Veracruz, but their map 364 is based on the records of Osgood (1909:158) and Dalquest (1953:143). I have examined all the specimens reported by the two authors last named and find that those from San Luis Potosí are P. boylii levipes.

The diagnosis and comparisons here presented of aztecus were based on specimens from Mirador in comparison with all the specimens of P. boylii from eastern México listed beyond. The largest specimens of P. boylii that I have examined are from Las Vigas, Veracruz, and localities within a radius of five kilometers thereof. Some measurements of these large specimens of P. boylii overlap those of P. aztecus but the two kinds of mice differ greatly in characters of the skull, in color, and in length of tail.

The specimens (three adults and three juveniles) from Huachinango, Puebla, are slightly darker than specimens from Mirador but do not differ otherwise. Of two specimens reported from Jalapa, Veracruz, by Osgood (1909:158), one (108547 USNM) agrees with specimens from Mirador in color and cranial characteristics and is P. aztecus, whereas the other (108548 USNM) is P. b. beatae.