OF THE MALTESE HILLS. 



491 



An especially noteworthy point is the great difference in texture 

 between this rock and the rocks with which it is associated. The 

 latter are very coarsely crystalline and extremely rich in horn- 

 blende. This rock, on the contrary, is of much finer texture, and 

 one of its chief constituents is epidote, a mineral of secondary origin. 

 The fibrous alteration-product already mentioned (sillimanite or 

 natrolite) chiefly characterizes those particular bands in the section 

 which give the foliated appearance to the specimen. 



There is no elongation of the quartz- and other constituent crys- 

 tals to indicate that the banding of the rock is due to stresses and 

 earth-movements, which in some cases are known to induce foliation 

 and schistose structure in eruptive rocks, and it seems by no means 

 improbable that this rock, in its earlier condition, possessed such 

 banded structure as now exists in it (PI. XIX. fig. 2), and that this 

 banding resulted from the accumulation of hornblende &c. in certain 

 planes *. These, I believe, were planes of stratification ; and I am 

 disposed to regard the rock provisionally as a highly altered sedi- 

 mentary deposit or an altered and bedded volcanic tuff. Indeed, 

 it seems very probable that in the earlier periods of the earth's 

 history, when sedimentary rocks must necessarily have formed a 

 far smaller proportion of the earth's crust than they do now, the 

 products of denudation were chiefly derived from rocks of an erup- 

 tive type, and deposits formed of such materials would approximate 

 more or less closely in mineral constitution to beds composed of 

 truly volcanic ejectamenta. The sedimentary rocks of later date 

 consist, in great part, of partially decomposed and triturated mate- 

 rials derived from the repeated partial decomposition and trituration 

 of rock, and there is, consequently, less probability now of the 

 sedimentary deposits resembling volcanic tuffs than there was in 

 Archaean times. 



The minute structure of the rock is not gneissic or schistose, and 

 its fissile character is dependent upon the layers of epidote, which 

 give it its banded appearance. 



Xo. 3. North Hill. Easternmost quarry on Nortli side. — This is a 

 coarsely crystalline rock apparently consisting of black hornblende 

 and flesh-coloured felspar, and in which traces of foliation are only 

 perceptible in large specimens. 



Under the microscope the rock is seen to consist of hornblende, 

 triclinic felspars, magnetite, and apatite. The felspars are mostly, 

 if not exclusively, triclinic, and are, as a rule, more or less altered 

 by the development either of felsitic microcrystalline matter, or by 

 minute scales, with here and there elongated negative crystals 

 sometimes filled with decomposition-products. In one of these 

 cavities a bubble is perceptible. All these results of alteration 

 follow, as a rule, the direction of twin-lamellaa, although at times 



* In accordance with what is stated in the concluding pages of this paper. I 

 may add that it is possible that such a rock might result from the crushing and 

 chemical alteration of a diorite ; and the proximity of diorite -would lend 

 support to this view. The microscopic evidence does not, however, appear to 

 favour such an hypothesis. — June 28, 1887. 



