Ilgperslkene Rock, 4G5 



oblate spheroidal shape. This rock is singularly durable ; seem- 

 ing to undergo no waste capable of generating a soil, either upon 

 the surface or at the feet of the mountains. Hence they pre- 

 sent an aspect of entire desolation, their whole visible declivity, 

 where they surround the lake, being brown and bare, with scarcely 

 a mark of vegetable life. 



The geological relations of this rock are ascertained, not only 

 by its intimate connexion with the greenstone, basalt, amygda- 

 loid, and sienite, that form the greatest part of the island, but by 

 its actual overlying position on beds of red sandstone, which are 

 fully described in Dr. M.'s papers before the Geological Society, 

 from which we have extracted these circumstances. Hyper- 

 sthene rock is therefore a member of the trap family ; and we 

 think there is reason to conclude that it is also of later origin 

 than the lias formation — a reason derived from the author's ac- 

 count of the other parts of the island. 



There are many varieties of structure and appearance ; but the 

 varieties of composition seem reducible to three : hypersthene 

 with greenish compact felspar, with dark purple ditto, and with 

 crystallized felspar approaching to glassy. The concretions vary 

 much in size : when large, the nature of the rock is easily ascer- 

 tained ; when minute, it is easily mistaken for greenstone ; since 

 the hypersthene may be confounded with hornblende, a mistake 

 which the author suspects to have been often made. The purple 

 felspar resembles that of Labrador, but is never iridescent. Con- 

 temporaneous veins traverse the rocks in many places, containing 

 very large concretions, and affording fine specimens of the several 

 mineral substances. It must also be added, that veins of oxidulous 

 iron ore are found in it ; and that crystals of this substance of 

 a tetrahedral form constitute in many cases an integrant part of 

 the rock. 



Jt is probable that a similar rock is the birth-place of the hy- 

 persthene and purple felspar of Labrador, and that it will be found 

 hereafter in many places where it has either been overlooked, 

 or confounded with common greenstone. It has generally hap- 

 pened in the history of mineralogical discovery, that a substance 

 has been found common when once it had been shown to exist. 

 We may instance the quartz rock of the same author, long over- 

 looked as an occasional substance of trifling extent, or confounded 

 with mica slate, and now proved to form a leading division of the 

 primitive rocks all over the globe, since it occurs in the moun- 

 tainous regions of India and southern Africa, occupying tracts of 

 great extent. 



A proof of this is given in the author's discovery of the same 



I lypersthene rock in Airdnamurchan, a spot frequently examined 



bv gnologisis. Here, as in Sky, it forms an extensive tract, hi- 



' Vol. 49. No. 2.'U). June 1817. <-' g thcrto 



