the replacing hornblende, there appears some little difference in the various aggregates—some (and these show no signs of diallage) being more "matted" in texture, if I may use the phrase, than others, and showing here and there a little of a non-doubly refracting mineral rather like serpentine. It can hardly be supposed that one part of the same mass of gabbro would be rich in olivine and another at no great distance entirely without it; hence I conclude that sometimes the olivine in a gabbro, instead of changing to serpentine, becomes converted into hornblende: the requisite silicate of lime must be supplied by the neighbouring felspar—a thing not impossible when we see how it is often penetrated by these new acicular crystals.
There are some dykes in this mass of gabbro. Soon after getting on it we come, as we clamber along the shore, to two small dykes, running rather irregularly, of a sharp-jointed, rather splintery, very compact dull-greyish or greenish rock, of a rather serpentinous aspect, the larger generally about 18 inches, and the smaller about 15 inches wide. I have had a slide cut from the former. It proves to be a basalt, chiefly consisting of small crystals of plagioclase and augite, both rather altered, and rather poor in magnetite. I suspect that much of the plagioclase is now a pseudomorph; we have the irregular low-tinted granular aspect in part of the field. A good deal of the augite has undergone change, and should, perhaps, rather be called uralite; but that the rock has been a basalt I have no doubt. There is a little apatite. A short distance further is a dyke from one to four yards wide, which forms a little headland. This is in parts very like an ordinary basalt or anamesite, with a glistening surface and weathering brown. Parts of it are rather porphyritic, having very white felspar crystals up to about 14 inch long. I have had slides cut from this; and on examination it proves to be a basalt, the plagioclase being fairly well preserved, the augite sometimes very characteristic, and a good deal of olivine with a granulated dusky aspect like (though on a smaller scale) that described above; there is also a fair quantity of magnetite. The most porphyritic variety of this rock, macroscopically, so closely resembles the most porphyritic variety of the "diorite" at Poltesco, that the specimens might easily be confused.
Beyond this I observed two other dykes:—one resembling a basalt, not more than a foot wide, in a little headland; the other some quarter of a mile further on, also like a basalt, very compact, splintery in fracture, much cracked and jointed. The above-described dykes lie well in the first mile of the gabbro. After following the shore for about that distance—a slow and rather laborious process—I was obliged, by want of time, to take a path along the low cliff above, as I was anxious to examine the greenstone of St. Keverne.
This I was not able to do as completely as I had hoped; for where from the map I had expected to come well into the mass (here coloured over a space nearly half a mile across), I was still on gabbro, pierced by frequent veins of greenstone; and I had not time
