'; ?>
click to enlarge
164 almost invariably broken into small bits and water-worn. They are indubitably derived from the erosion of the underlying Pyrotherium beds. 3. Aside from bones derived from the underlying beds, I found no fossils in this. I believe it to be a freshwater series contemporary with or immediately preceding the earliest transgression of the Patagoniano sea. 4,5. These beds are apparently con- temporaneous. Both series contain very numerous fossils, mammals and birds, of Pyrotherium or Descado age. They are of slightly different facies. The only difference noted in my very rapid study was that Pyrotherium itself is definitely more common in (perhaps even confined to) the channel facies. 6. The surface of 6 is, of course, covered with fragments washed down from the overlying Descado beds. There are definitely some fragmentary bones in place in 6, but we found no identifiable fossils surely in place. The unconformity, 4, 5-6, suggests that 6 is definitely older, although not in itself disproving the possibility that the disconformity is local + the age difference slight. The discoveries 165 in section DB, however, strongly suggest that 6 belongs to the Notostylops age, Casamayor. [The above section and notes have been made without any reference to the work of Loomis and are subject to revision by any facts overlooked in our rapid study that he may have found in his very much longer and more detailed work at the same locality.] Although it must always be possible to pick up a few good specimens at this locality, it definitely will not repay long- continued collecting at present. Ameghino worked it. Loomis obtained a magnificent collection which must have exhausted the beds temporarily and have made an equal crop impossible without many years of erosion. Riggs also went over it thoroughly + gathered the crop developed since Loomis. Other geologists visit it almost every year and pick up all the obvious things. I found about 25 identifiable specimens, and probably exhaustive collection this year would not more than double that number. Almost all