definitely the imbedding material which ingulfed and destroyed the living trees, and in which the petrifactions are now preserved. It is, as a rule, a volcanic conglomerate, or more properly a volcanic agglomerate. Both the matrix and the imbedded particles are truly volcanic, the latter varying from dust particles through all sizes up to those of a ton or more in weight. That the material has been accumulated under the partial influence of water or liquid conditions is evident from the more or less perfect stratification which generally pervades it. But the fragments are too angular, brecciated, to have suffered transportation and deposition as subaqueous or as ordinary river deposits. The attractive and satisfactory explanation of conglomerate formation in the Utah plateau region, as given by Captain Button, I do not think is here applicable.
From what is said in regard to the series of forest growths, and also from the evident thinness of the layers of débris, it is seen that there have been many successive sheets of the material laid down at the same place. In some cases and at certain places a true lava flow has spread over the surface, but the lava ledges can at points be seen to shade into the brecciated layers. While not believing that the great mass of breccia, covering perhaps hundreds of square miles, has been literally ejected from volcanoes, as has been held in regard to such formations, I am of the opinion that the accumulation of it is the direct and immediate result of such eruption.
Extruded lava from any source, not being perfectly liquid, would cool with an irregular surface, and terminate in precipitous ledges. This unevenness of surface, combined with the original slope that must have existed to permit any flow, would soon cause the whole area involved to be abundantly floored with volcanic fragments of all sizes. During subsequent eruptions these fragments would be swept along by and with the liquid matter, commingled with dense showers of ejected material, amid heavy flows of water from accompanying rains and perhaps melting snows, to be deposited in layers at varying distances from the centers of eruption, the condition in which it is now found. Most of the material of which the agglomerate is composed I believe to have come by the ordinary process of weathering of previously erupted rocks, and then to have been commingled with finer ejected material and distributed by the floods which accompanied some if not all the outflows. The interstratified beds of varying degrees of fineness are the results of less tumultuous periods.
Such explanation involves the necessity for many centers of eruption in the park region, for the agglomerate is of wide extent, and it could not be formed at great distances from these centers.