stantly augmenting and at the same time undergoing decay. This is finally expressed in the deep black color of the stratum, by the carbonized fragments of marine algae, driftwood and even of bones, showing that within this zone there are developed precisely those conditions which would be productive of gases in considerable volume.
It is this last explanation which affords the chief basis of a tentative hypothesis respecting the origin of the gases producing the conflagrations, though it is also highly probable that other volumes of gas originated at a greater depth in a buried marsh, or in silt deposits which were subsequently overlaid by a pebbly beach.
This phenomenon, while peculiarly interesting in itself, serves as a means of explaining the possible origin of many obscure forest fires for which it has hitherto been impossible to find an adequate explanation, and in considering this important aspect of the question we are not to overlook the possibility of accounting for fires which have occurred in past geological ages, as well as those of recent date.
In 1905, Arthur Hollick directed attention to the presence of charred wood in the Cretaceous deposits at Kreischerville, Staten Island, New York, and drew the inference that since man was not in existence at that time, the fire must have been due to some natural agency, probably lightning. This explanation, however, was not regarded by him as wholly satisfactory, and it was adopted tentatively because of the absence of positive testimony in any other direction, and also because the occurrence of fires in widely separated localities of approximately the same geological age could not be accounted for through the medium of such an agency.[1] In a more recent communication on this subject,[2] the same author observes that some of the fragments of burned wood are charred on the outside only, while other smaller fragments are completely charred throughout. "These latter occur in greatest abundance in connection with layers or seams of yellowish, sandy clay. The prevailing colors of the Cretaceous sands and clays throughout this locality are white and gray, while the yellow layers are of quite limited extent and appear to have been burned or baked. It seems therefore reasonable to infer from this association of materials, that the charred wood was not deposited with the clay in the condition of charred wood, but that it was fresh material at the time of deposition and was subsequently burned in place, thus baking the enclosing clay."
"A careful study of the Kreischerville deposits indicates very clearly that the original conditions of deposition must have been strikingly similar to those described as existing at the Kittery Point Beach. The layers of vegetable debris and sand, intercalated in the clays are comparable to the sandy layer of black, organic débris
