MATTHEW, PATAGONIA AND PAMPAS CENOZOIC 
153 
expressed views, as well as with Ortmann, Scott and Hatcher, in regarding 
them as the marine and fresh-water facies of a single great formation. So 
far as the age of the marine facies is concerned, he points out the discrepancy 
in the conclusions of Cossmann, von Ihering, Ortmann, Ameghino and 
Wilckens as to the age indicated by its fauna, and concludes that the marine 
fauna is a rather uncertain guide as to the age of the formation. The 
reviewer ventures to express a corresponding skepticism as regards the 
marine facies of the Notostvlops beds. The correlation is not much more 
satisfactory when obtained through the terrestrial fauna, large and well 
known as this is. This fauna is in great part composed of new groups of 
mammals, which are not directly derivable from those of the Pyrotherium 
beds; other groups have broadened out or specialized so far between the 
two epochs as to show that a long time gap intervenes. The intermediate 
stages recognized by Ameghino between the two are regarded by Roth as 
not demonstrably more than local facies of the Santa Cruzian fauna. 
There are at most three sub-divisions of the Patagonian tuffs. The 
lowest member, the Tecka beds, contains a limited mammalian fauna of 
older facies, as shown by the presence of Archceohyrax of the Pyrotherium 
beds, and the continued absence of rodents. The middle horizon includes 
the main mass of the marine Patagonian, in which Santa Cruz mammals 
are found locally, mixed with the marine fauna. There are also consider¬ 
able fresh-water mammaliferous beds in this horizon. The main body of 
the epicontinental Santa Cruz formation lying to the southward, overlies 
the marine beds, according to the observations of Carlos Ameghino, and if 
so, constitutes the uppermost member of the formation. 
Much weight is laid by Roth, as also by Ameghino, upon the evidence 
obtainable from the rodents in correlating the Santa Cruz fauna. In Europe, 
rodents first appear in the Lower Eocene (Wasatch and Suessonian), in 
the Argentine their first appearance is in the Patagonian tuffs . 1 All the 
Santa Cruz rodents are highly specialized forms; the primitive groups of 
the European Eocene do not appear at all. On the other hand, no modern 
genera occur in the Santa Cruz, while in Europe a large percentage of the 
Miocene genera are still living, and many living genera are found even in 
the Oligocene. The author concludes that this entire absence of living 
genera indicates an age not later than Oligocene for the Santa Cruz 
rodentia. 
The reviewer would agree as to the value of the rodents in this problem, 
but would be more inclined to weigh their actual degree of diversity as a 
whole from the modern rodents, than the rather nominal character of per- 
1 Ameghino, however, records Cephalomys , an unquestionable rodent, and a specialized 
Hystricomorph at that, from the Pyrotherium Beds. 
