122 



University of California. 



[Vol. 3. 



refraction and uniaxial and positive character. Each of the 

 rounded areas in which it occurs contains a number of inter- 

 locking allotriomorphic grains. The red color is due to finely 

 divided ferric oxide, which occurs along the boundaries of the 

 cluster and between the individual grains. This quartz is evi- 

 dently a secondary filling of cavities, but whether these cavities are 

 original or are formed by differential solution is doubtful. 



ROCKS OP THE CLARNO EOCENE. 



General Classification . — The Clarno is essentially a volcanic 

 formation. The clastic rocks are tuffs and agglomerates of 

 widely varying texture, and massive lavas also constitute an 

 important part of the series. While the writer is unable to give 

 in detail the succession of the various rock types represented, a 

 general idea may be gained by considering the sections at the 

 localities where these rocks are exposed. 



In Hald's Canon, near the town of Mitchell, the section con- 

 sists of a considerable thickness of hornblende and pyroxene ande- 

 site which seem immediately to overlie the supposed Cretaceous. 

 At Clarno 's Ferry the lowest exposed beds are andesite tuffs, 

 containing fragments of both hornblende and pyroxene andesite. 

 Above these tuffs are andesite lava, and the Eocene section is 

 topped by rhyolitic lavas and tuffs, overlaid by the red beds of 

 the John Day Miocene. In the Cherry Creek area, the andesitic 

 tuffs, with leaves similar to those found near the base of the 

 Clarno' s Ferry section, are overlaid by pyroxene andesite, quart 

 basalt, and rhyolite. At Bridge Creek, the leaf beds just 

 beneath the John Day series are underlaid by rhyolite, while 

 quartz-basalt and pyroxene andesite occur lower down the 

 canon, and, it appears, stratigraphically beneath the rhyolite. 



Comparing the various sections, we find the succession to 

 be, speaking generally: (1) Andesite, (2) Quartz-basalt, (3) 

 Rhyolite. To work out the succession in detail would involve 

 more careful field study than Dr. Merriam or the writer has been 

 able to give to the subject. The pyroxene andesites are probably 

 later than the hornblende andesites. 



Pyroxene Andesites. — The most representative specimens of 

 this type of lava were collected in Hald's Canon. The material 



