66 Proceedings of the Royal Physical Society. 
above minerals are associated in these clusters. There is much 
saponite in small and loosely arranged flakes. This is not 
in small druses, but in irregular patches, as if it had replaced 
augite. Lath-shaped crystals of labradorite penetrate these 
patches, showing that this mineral had been protero-genetic 
to the augite. Analcime in somewhat large crystals occurs 
as an imbedded ingredient.” 
“Though not for a moment calling in doubt the fact that the 
piece of rock sectioned came from Rockall, I must express the 
very strongest doubts as to this being the usual, or main, 
rock of that extended reef. Such a mass as that would stand 
above the water in high, cliffy bluffs, with deep water up to 
the very edge of the precipitous sides, It seems to me much 
more probable that the fragment of rock sent had been a 
portion of a number of volcanic dykes which had cut the 
true rock in all directions, and, by their greater hardness, 
had saved it from utter disintegration. Moreover, the 
specimen sent was itself of an unsatisfactory character. It 
did not show properly fractured edges, these being rounded 
from more or less atmospheric or marine rubbing. It looked 
like a fragment which had been dislodged out of a rent in a 
stone.” 
Early in 1894 some talk was made of erecting a light- 
house upon Rockall, and, in a letter received by me on 28th 
April 1894, from Mr D. A. Stevenson,—Secretary’s office 
of the Board in Edinburgh,—that idea for the time being 
appears to have been dismissed, after all possible inquiries 
had been made. Mr Stevenson wrote: “It is evident that 
there is not much information forthcoming from those who 
have actually landed on the rock.” 
The difficulties of storage, of laying a cable, supplying 
victuals to workmen excavating for foundations, first for 
storehouses and workshops, and secondly for the lighthouse 
itself, distance out in the Atlantic, and the incompleteness 
of charts and soundings, for the present, at least, render 
the prospect unlikely. 
Until a more careful survey both of the rock itself and 
its capabilities, and of the surrounding seas, be completed, it 
is not likely that any steps can be advanced in this direction, 
