Plagioclase, augite, olivine, magnetite, and, in some cases, sanidine—possibly titaniferous iron—and in some, if not in all, a slight residuum of glass. These lavas must therefore be regarded as plagioclase-basalts, or occasionally as olivine-basalts. The plagioclase crystals vary greatly in size, some being mere microliths while others are over the eighth of an inch in length. They show the characteristic twin lamellation by polarized light, but the lamellæ are often very irregular as regards their boundaries. The sections of the crystals themselves are also frequently bounded by irregular outlines, but they often show internally delicate zonal markings, as indicated in Fig. [1],[22] which correspond with the outlines of perfectly developed crystals. The inclosures in the larger plagioclastic felspars consist for the most part either of brownish glass, containing fine dark granular matter—probably magnetite, which often renders them opaque,—or of matter similar to that which constitutes the groundmass of the surrounding rock. These stone and glass cavities are very numerous and most irregular in outline, as shown in Figs. [1] and [2]. They appear, however, to be elongated generally in the direction of the planes of composition of the twin lamellæ. Zirkel has noted the plentiful occurrence of these glass inclosures in the felspar crystals and fragments of crystals which partly constitute the volcanic sands of Etna, in which he has also detected the presence of numerous isolated particles of brownish glass.[23] The felspar microliths, which constitute so large a proportion of the ground-mass in the Etna lavas, are in most instances probably triclinic. Monoclinic felspar does, however, occur in some of these rocks; but the difficulty of ascertaining the precise character of microliths renders it unsafe to speculate on the amount of sanidine which may be present. Some crystals, such as that shown in the centre of Fig. [2], appear at first sight to be sanidine, twinned on the Carlsbad type, but closer inspection often demonstrates the presence of other and very delicate twin lamellæ.
The augite in these lavas sometimes occurs in well-formed crystals of a green or brown colour, and often shows the characteristic cleavage very well, especially in the augite crystals of the lava of the Boccarelle del Fuoco, erupted in 1535. A small crystal of green augite is represented at the bottom of Fig. [1]. Augite, however, appears to be more plentiful in the rocks in the form of small roundish grains.
Olivine is of very common occurrence in the Etna lavas, mostly in round or irregularly shaped grains, but also in crystals which usually exhibit rounded angles.
A specimen of lava from Salto di Pulichello, erupted in 1603, gave well-developed examples of the presence of olivine, and also of plagioclase. The ground mass was found to consist of felspar microliths, and grains of olivine, augite, and magnetite, with some interstitial glass.
Magnetite is present in all of the lavas here described. It occurs both in octahedral crystals and in the form of irregular grains and fine dust. To the presence of this substance much of the opacity of thin sections of the Etna lavas is due.
Titaniferous iron may also be present. One small crystal in the lava of 1535 appeared to show a somewhat characteristic form, but although much of the black opaque matter has undergone decomposition, I have failed to detect any of the white or greyish alteration product which characterises titaniferous iron, and in the absence of this, of definite crystalline form, and of chemical analysis, it seems better to speak of this mineral with reserve, although titanium is very probably present, since much magnetite is known to be titaniferous.
The vitreous matter which occurs in these lavas is principally present in the form of inclosures in the felspar, and, sometimes, the augite and olivine crystals previously described. Its occurrence in the groundmass of these rocks has also been alluded to. In this interstitial condition its amount is usually very small—a fact already pointed out by Zirkel.
I have unfortunately had no opportunity of examining the volcanic sands and ashes of Etna, but Zirkel's description of them seems to indicate their close mineralogical relation to lavas erupted in this district, with one exception, as pointed out by Rosenbusch,[24] namely, that he makes no mention of the occurrence of olivine in these ejectamenta.
Reference to the Figures [1] and [2] will suffice to show how close a relationship in mineral constitution exists between these two lavas, separated in the dates of their eruption by an interval of over two thousand years.