The specimens examined are referred, tentatively, to _Geomys personatus personatus_ on geographic grounds. They average smaller in all measurements than _personatus_ (but are larger than _G. p. megapotamus_), do not have the sagittal crest that usually is present in _personatus_, and the shape of the pterygoid bones is distinctive. In _personatus_ and _megapotamus_ the ventral border of the pterygoids (in lateral view) is convex instead of nearly straight as in specimens from the barrier beach. The specimens recorded here are all that are known of _G. personatus_ (see account of _G. tropicalis_) from México. _Measurements._--Average and extreme external measurements of five females from 73 miles south of Washington Beach are as follows: 266.8 (263-271); 94.8 (91-98); 34 (33-35). Cranial measurements of two males (89038, 89032) and average and extremes of five females are respectively: basal length, 49.1, 46.6, 45.9 (44.2-46.8); basilar length, 42.9, 40.0, 39.8 (38.0-40.8); zygomatic breadth, 29.6, 28.3, 28.0 (25.7-29.9); squamosal breadth, 27.8, 25.9, 26.2 (23.8-25.4); interorbital constriction, 7.4, 6.9, 7.3 (6.7-7.8); alveolar length of maxillary tooth-row, 10.3, 9.2, 9.4 (9.1-9.7). _Records of occurrence._--Specimens examined, 17: 35 mi. SSE Matamoros, 8; 33 mi. S Washington Beach, 1; 73 mi. S Washington Beach, 8. Additional record: 4 mi. S Washington Beach (Selander _et al._, 1962:335--possibly fragmentary skeletal remains never catalogued in any research collection). =Geomys tropicalis= Goldman Tropical Pocket Gopher 1915. _Geomys personatus tropicalis_ Goldman, Proc. Biol. Soc. Washington, 28:134, June 29, type from Altamira, Tamaulipas. _Distribution in Tamaulipas._--Known only from vicinity of type locality, in southeastern part of state. _Geomys tropicalis_ was named as a subspecies of _G. personatus_ in 1915 by E. A. Goldman. To my knowledge, no one other than Goldman has critically studied specimens of this pocket gopher, nor have specimens other than those listed in the original description been reported up to now. In 1953, Gerd H. Heinrich collected a series of 19 individuals one mile south of Altamira. These specimens were compared (by E. R. Hall in March, 1962) with the holotype and paratypes of _G. p. tropicalis_ and were found to be indistinguishable. Careful comparisons of the specimens from one mile south of Altamira with topotypes of _G. personatus personatus_ (and specimens of other subspecies) indicate that _tropicalis_ differs from _personatus_ in a number of important characters, some of which _tropicalis_ shares with _Geomys arenarius_ of the Rio Grande Valley and adjacent areas in Texas, New Mexico, and Chihuahua (see Table 2). TABLE 2.--DIFFERENCES BETWEEN THREE SPECIES OF GEOMYS.