724 G. SHERBURNE ROGERS 
lenses of the white sandy tuffaceous material described above may 
represent the ancient stream courses. These lenses are numerous 
and generally of small size; in many cases they lie at a small angle 
across the strata and are themselves always strikingly cross-bedded 
(see Figs. 3 and 4). It would not be expected that recent erosion of 
the badland type would lay bare an old river course with its mean- 
ders for any great distance; the recent gulleys would cut across at 
all angles and the exposures of the old alluvium would pinch out 
abruptly. The shorter lenses would then represent an approximate 
cross-section of the old channel, while the longer ones might be 
interpreted as more or less complete approaches to a longitudinal 
section. 
APPLICABILITY OF THESE PRINCIPLES 
Lack of exactness in our present knowledge.—The application of 
the principles above outlined is naturally limited to certain forma- 
tions, and is at the present time hampered by our comparatively 
small knowledge of the petrology of clastic rocks. The lack of 
detailed microscopic study is making itself felt in the deplorable 
looseness of our definitions of the types. A committee on the 
nomenclature of the more common types of sedimentary rocks was 
recently appointed in the United States Geological Survey, and in 
the course of a rather exhaustive search through the literature in an 
endeavor to find some basis for standardizing and delimiting the 
types in common use encountered a surprising discordance of 
opinion. Shale, clay, graywacke, arkose, argillite, etc., are differ- 
ently defined in different books and in some cases not defined at all. 
With so little literature on the exact composition of sedimentary 
rocks and with the very basis of the lithology of these rocks in this 
condition it is difficult to say how great an aid microscopic work 
will prove in solving the problems of correlation or in reconstructing 
the conditions under which a sediment was laid down. At the 
present time it would seem that in many, if not most cases, such a 
study would have no immediate and practical value, but on the 
other hand, there are certainly many problems in which it might be 
used with direct advantage, as in the instance given above. 
The lack of exactness in our present knowledge may be further 
illustrated by the following example, which represents another and 
