GEOLOGY: J. BARRELL 
497 
tion prevails in British opinion to the present day, though it is true that 
Macnair and Reid^ reverted in 1896 to the hypothesis of marine origin, 
and Goodchild in 1904,^ pressing in the other direction, urged ''that the 
whole of this vast formation was accumulated under continental con- 
ditions, partly in large inland lakes, partly as torrential deposits, partly 
as old desert sands, and partly as the result of extensive volcanic ac- 
tion." In 1908 Walther published a volume which consisted of a gen- 
eral review of historical geology. In the section touching on the Old 
Red Sandstone, he held that the lakes of the British geologists were in 
reality desert basins which ''possessed a hot desert climate whose dry 
periods were broken only seldom by the downpour of thunder storms."^ 
In other places Kayser and Walther have expressed the view that the 
Old Red Sandstones were deposited in lagoons not far from the sea, the 
water being thought of as occasionally replenished by the sea and the 
deposition as taking place in lagoons and as dunes on their margins. 
Thus we see that a great variety of conclusions, amounting in fact 
to fiat contradictions, have been expressed and are still entertained by 
geologists in regard to the nature of the habitat of the early vertebrate 
faunas whose remains are found in the formations of the Old Red Sand- 
stone. Yet this is to the geologist and biologist a most important prob- 
lem, especially as it may be shown that the conditions of the environ- 
ment held important causal relations to the rise of amphibians. 
The differences in conclusions are not due to differences of opinion 
in regard to the actual facts of stratigraphy, but to radical differences 
in the interpretation of those facts. At the present stage of investiga- 
tion it is of the first importance therefore to determine the criteria for 
interpretation and their degree of validity. It is that subject especially 
to which the writer has given attention in previous investigations. The 
present conclusions rest therefore upon the following papers. 
In November, 1906, the writer, introduced by Professor Davis, gave a 
paper before the National Academy of Sciences entitled 'Continental 
Sedimentation with Applications to Geological Climates and Geogra- 
phy.' The articles on this general subject were published in extenso in 
the Journal of Geology in 1906 and 1908.^ In 1913 and 1914 articles 
were published in the American Journal of Science on the formations 
in the eastern United States which correspond to the Old Red Sand- 
stone system of the British Isles. ^ In the present article, therefore, the 
establishment of the criteria on which the conclusions rest may be 
omitted and the resulting interpretation directly given.^ 
This interpretation leads to the conclusion that the deposits which 
make up the Old Red Sandstone, although they undoubtedly contain 
