But let us try the tree-ferns and cycads of the coal beds of the "older" rocks. In northern regions they are not found "later" than the Triassic and Jurassic, and doubtless the same holds good of the rocks in the Tropics, where the modern species now live in fair abundance. But how did they come to shift to the Tropics so many millions of years before the palms, etc., of the Tertiaries thought it time to do the same? The climate had not changed a bit: how did they come to scent the coming "Glacial Age" so much earlier than their more highly organized fellows? The "Challenger" expedition found some Cyathophylloid corals now building reefs at the bottom of our modern ocean. The geologists had already assigned =the last= of them to the Carboniferous and Permian rocks with the idea that they were extinct. But where have these fellows kept themselves during all the intervening ages while the continents were deep under the ocean time and time again? or why are not the rocks containing their fossils as "recent" as any deposits on the globe? And so I might go on. There is hardly a tribe found in the "older" rocks which does not have its living representatives of to-day, and with, I believe, a fair proportion of the species identical; though in hundreds, perhaps thousands, of cases these species, genera, or even whole tribes, have somehow skipped all the intervening formations. But let us drop this method of studying our subject, and look at it from a slightly different standpoint. Thus Dana[50] says that: "The absence of Lamellibranchs in the Middle Cambrian, although present in both Lower and Upper, means =the absence of fossils from the rocks, not of species from the faunas=." He puts this in italics by way of emphasis, for it is certainly a reasonable idea, and as A. R. Wallace says, "no one =now= doubts that where any type appears in two remote periods it must have been in existence during the whole intervening period, although we may have no record of it."[51] But what would be the result if we only extend this idea to its logical conclusion? It seems to be an effort to avoid one of the absurdities of the onion-coat theory, without, however, discarding that theory altogether. In speaking of some corals and crinoids of the Devonian which "were absent" from some of the divisions of this formation because the conditions of the seas about New York "were unfavorable," Dana says that "they were back when the seas were again of sufficient purity."[52] In his review of these formations he enlarges on this subject: