Page:The Journal of geology (IA journalofgeology21894univers).pdf/275

large number are composed of only one or two cells. The annual rings are rather indistinct, yet can be made out.

"As far as I now know, only two species of Araucarioxylon have been described from the United States, A. Arizonicum Kn., from the Triassic or Lower Jurassic of New Mexico and Arizona, and A. Virginianum Kn., supposed at first to belong to the Potomac formation, but now known to be from the Trias of Virginia. These species differ markedly from the one under discussion. With the A. Arizonicum it has almost no points in common, while it differs from the A. Virginianum in important particulars."

The only sections of the fossil wood that have yet been made were cut from a specimen taken from the cycad bed proper and not from the principal fossil forest, but it often happens that only one species can be found in such a forest. It is therefore probable that the same structure would be shown by other specimens. I confess to a little surprise at finding that this structure represents the Araucarian rather than the Sequoian type of conifers, since, in the east at least, these two types characterize the Trias and Potomac respectively, no Araucarian specimens having been found in the Potomac and no Sequoian specimens in the Trias. And generally the Araucarian type is more ancient. This evidence therefore points to a lower instead of a higher horizon.

I have made a somewhat careful study of the specimens from the plant bed above Evans Quarry, and have asked Professor Knowlton to assist me, his experience in recently editing Lesquereux's Flora of the Dakota Group having familiarized him with the forms of that age. The result of our joint investigation may be summed up as follows:

The specimens are few and fragmentary, and the only species that can be even approximately determined are: