42 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
of wonderfully massive structure, steep, bare, and clean: they frequently 
show cross-bedding, but the texture of the sands is so uniform that 
there is little or no weathering along the bedding planes. The cliffs 
retreat by the detachment of huge slabs on vertical joint planes, but the 
amount of talus seems very small in proportion to the great height of 
the cliffs. Hanging lateral valleys open at various heights in the main 
valley walls ; and above the level of the higher lateral valleys the cliffs 
rise in rounded domes, which seem to record a more mature stage of 
erosion during an earlier cycle, when the land stood lower than at 
present. 
The rolling plateau country in which the canyon is incised was 
described to us as affording abundant pasture and timber. <A zigzag 
path has recently been constructed on one of the more battered parts of 
the canyon wall: part of the path consists of tree-trunks held against 
the rock face by iron clamps and brackets. Cattle are here driven up 
to the highlands for summer pasture. A wire rope has lately been 
stretched from a bold cliff-top down into the valley, to serve for lowering 
boards and shingles from the upper forest. The height of the cliffs is 
so much greater than can be properly estimated that I refrain from 
indicating it in feet ; but the rough units of measure should be thousands 
rather than hundreds. 
The Fresh-water Tertiaries. 
Previous Statements. — During the past summer I had opportunity 
of examining with some care several characteristic sections of the fresh- 
water Tertiary formations which occupy so large an area in the Cordil- 
leran region. The localities to be described here are those of the Eocene 
beds exposed in the Pink cliffs of the Paunsagunt plateau and there- 
abouts in southern Utah, and the Green river beds near the town of that 
name in Wyoming. The object of this study was to consider on the 
ground whether the theory of the unqualifiedly lacustrine origin of these 
formations, presented in the various reports on the geology of the region, 
would command entire acceptance. The lacustrine theory has been, 
until recently, practically undisputed for thirty or more years, but a 
number of essays have now been published in which the capacity of 
agerading rivers and of winds to form extensive non-marine deposits has 
been recognized, and in which the effort has been made to discriminate 
carefully between fresh-water formations of different origins. In a 
number of these essays, especially in those by Matthews, Hatcher, 
