Page:The Journal of geology (IA journalofgeology11893univers).pdf/25

twinning lamellae. On the other hand, the pinacoidal parting is entirely absent in cases where twinning lamellae are present. Consequently not much dependence can be placed upon this constituent as a means of distinguishing between gabbros and diabases. The former rocks are evidently related to the latter, whose typically granular, holocrystalline forms they are. Irving,^[1] in his work on the geology of the Keweenawan series in Michigan, Wisconsin, and Minnesota, was compelled to make use of coarseness of grain as a means of distinguishing between diabases and gabbros, both of which were thought by him to occur as flows. "It is evident," he writes, "that my observations on these north Wisconsin gabbros bear out the conclusions reached by certain European lithologists, as to the subordinate importance of the foliated condition of augite, by which gabbro is ordinarily separated from diabase, of which it would seem to be merely a phase. Nevertheless, the name is here retained, not only because most of our rock is very close to the typical European gabbros, but more especially because it is so sharply contrasted with the typical Keweenawan diabase that a separate name seems necessary." And again, when speaking of the diabases, he says,^[2] "Although grading through coarser kinds into the coarse olivine-gabbros, the fine-grained rocks here considered deserve a place by themselves. The gradation into the coarser kinds has never been observed in any one bed, and they are very strongly marked by their external characteristics, both in the fresh and altered states."

The prime distinction between the two classes of rocks is, then, one based upon structure and not upon the difference between the augitic and diallagic nature of its pyroxenic constituent. The structure of the most typical gabbros was recognized by most geologists to be granitic and that of the diabases as ophitic. Professor Judd^[3] proposed to restrict the name