452 University of California Publications in Geology [Vol. H 



Comparison of the Barstow fauna with that of the Rattlesnake 

 and Thousand Creek is necessarily limited to a few groups, owing 

 to the small representation of comparable types in the later faunas. 

 The horses of the Rattlesnake and Thousand Creek comprise only 

 advanced types included in the genera Pliohippus and Neohipparion, 

 in contrast to the abundant Merychippus fauna of the Barstow. The 

 camels of the Thousand Creek and Rattlesnake are in general larger 

 forms than those of the Barstow. The advanced types of antelopes 

 of the Thousand Creek fauna, represented by Ilingoceros and 

 Sphcnophalos, are much more progressive than Merycodus of the 

 Barstow. They are possibly derivatives from the Merycodus group, 

 which is not known in beds of the Thousand Creek stage. 



The almost total specific distinctness of the Barstow and Ricardo 

 faunas, taken with the wide difference of genera in groups with some- 

 what similar relations to their environment, makes it impossible to 

 conceive of the two faunas as having existed contemporaneously in 

 regions of quite similar topography only a few miles apart during 

 the time required to deposit the many hundreds of feet of strata in 

 which they occur. Comparison of the Barstow and Ricardo faunas 

 shows that in nearly every comparable group in which there is a 

 noticeable difference the Ricardo forms are more advanced. In the 

 Canidae the typical Barstow 'Tephrocyon disappears in the Ricardo. 

 The aelurodons of the Barstow are, where comparable, less specialized 

 than those of the Ricardo. The protohippine Mcrychippus forms with 

 a possible rare Protohippus or Pliohippus of the Barstow give place 

 to specialized Tlipparion and Pliohippus, with no forms as primitive 

 as Merychippus. The camels of the Ricardo include larger forms than 

 those of Barstow. Dromomeryx of the Barstow is not known from the 

 type section of the Ricardo, and the Ricardo Merycodus appears 

 somewhat more specialized than that of the Barstow. 



The difference between the Barstow and Ricardo faunas can 

 scarcely represent a time interval amounting to less than one-third 

 of a geological period as faunal changes are ordinarily interpreted. 



The exact relation of the Barstow fauna to that of the Cedar 

 Mountain beds 21 is not entirely clear. The Cedar Mountain carnivores 

 include a Tephrocyon apparently identical with T. kelloggi of the 

 Virgin Valley, while the T. temerarius type of the Barstow fauna is 

 absent. One large Aelurodon-like form of the Cedar Mountain does 



2i Merriam, J. C, Tertiary vertebrate fauna from the Cedar Mountain Eegion 

 of Western Nevada, Univ. Calif. Publ., Bull. Dept. Geol., vol. 9, pp. 161-198, 1916. 



