FOSSIL VERTEBRATES FROM SAN PEDRO VALLEY—GAZIN 483 
ment is the presence in the collection obtained from the above 
mastodont quarry of remains of true pes and apparently Sylwilagus, 
rather than Hypolagus. 
The absence of Vannippus from the Curtis ranch occurrence, geo- 
graphically so close to that of Benson, can scarcely be attributed to 
collecting chances or methods, as ine material of a fragmentary 
nature was observed at many places in the two areas, and none in the 
Curtis ranch embayment were of the Vannippus type. However, teeth 
of the Curtis ranch type of Hquus were found by Bryan near a powder 
mill a couple of miles south of the Benson locality, opposite the town 
of St. David. These can scarcely be said to occur at the Benson level, 
but indicate rather the occurrence of deposits of Curtis ranch age on 
the west side of the valley, undoubtedly at a level stratigraphically 
higher than that nearer Benson. 
It has been argued that there cannot be an appreciable time interval 
between the Curtis ranch and Benson occurrences because they occupy 
similar stratigraphic positions in the San Pedro Valley sequence. It 
is true that the interval between them is probably not great, but the 
distance and topographic features separating the two localities pre- 
clude any but speculative geologic correlation, nor is there any need 
to regard the sequence as having accumulated in a relatively short 
time. 
To account for the differences between the faunas of the two locali- 
ties and at the same time preserve what appears to be a nearly equiva- 
lent stratigraphic position, Gidley supposed that the Curtis ranch 
fauna as known did not evolve from that represented at Benson. 
This is in part obvious but is not necessarily proved for certain forms 
that may have come through, such as the larger cat and camel, the 
antilocaprid, kangaroo rat, pocket gophers, Baiomys, grasshopper 
mice, and cotton rats. Were the Benson fauna more completely 
known it seems likely that an even larger portion of the Curtis ranch 
ancestry would be included. In line with this it may be noted that 
glyptodont material was found in association with Nannippus and 
Plesippus from what may be regarded as the Benson stage in the 
nearby Gila Valley. 
The forms that seem most likely to have been involved in important 
migrations during the interval between Benson and Curtis ranch 
time are the horses, deer, mastodonts, and lagomorphs, these together 
giving the more obvious criteria for age assignment. 
Recently a relatively large fauna has been reported by Barbour and 
Schultz’ from deposits in Morrill County, Nebr., regarded as early 
Pleistocene. This, the Broadwater fauna, has not been reported in 
full, and only a tentative list of the forms represented has been pub- 
7H. H. Barbour and C. B. Schultz, Amer. Mus. Nov., no. 942, pp. 1-10, 1937. 
469725—42—_2 
