1894. 



PROCEEDINGS OF THE NATIONAL MUSEUM. 



657 



be determined macroscoijically, made up the entire mass of the rock. 

 Its appearance may perhaps be better understood by comparing it to a 

 conglomerate in which the large pyroxenic portions represent pebbles 

 and the hornblendes the interstitial cement. The rock is very massive, 

 and lam inclined to believe an eruptive, though definite proof is lack- 

 ing. Eruptive or otherwise, it seems to cover a very limited area. The 

 country rock is gneiss with a pronounced banded or foliated structure. 

 Passing along through the woods one comes suddenly upon an exposure 

 of this rock utterly different in mineral nature and structure, and occu- 

 pying an area so far as exposed scarcely a hundred feet in diameter. 

 Like so many other undoubted eruptives in the region, it occurred in 

 the form of what we after a time dropped into the habit of facetiously 

 calling imstules. Inasmuch as I have never seen any such sudden 

 transitions in gneissic rocks, but have observed basic undoubted 

 eruptives occurring in just this manner, I am naturally inclined to 

 regard this also as an eruptive, though contacts are wholly obscured. 

 It is apparently the deeper lying portion of an old and very small 

 volcanic neck. 



Thin sections under the microscope show the rock to be made up 

 wholly of large, irregularly outlined plates of hypersthene, pleochroic 

 in faint reddish and brownish colors, and a light-green hornblende, as 

 above indicated. These, with a scattering of opaque granules of iron 

 ore, comprise the entire list of recognizable constituents. The rock is 

 beautifully fresh and unaltered. The crystallization of the two chief 

 constituents must have been nearly contemporaneous. The hyi^ers- 

 thenes never show good idiomorphic forms, but the borders are irregu- 

 larly indented by the smaller hornblendes, which are also found in quite 

 perfectly outlined forms wholly inclosed by the hypersthenes. As a 

 rule, as noted above, these hornblendes occupy tlie position of a bind- 

 ing constituent, but at times both hornblendes and hypersthenes occur 

 intimately associated in small, imperfect granular forms. Of the two 

 minerals the hornblendes are the better developed. These show on 

 cleavage plates parallel to co P <^ extinction angles as high as 14°. An 

 analysis yielded results as below, all the iron being determined as 



SiO, . 

 AI2O3 

 Fe.,03 

 CaO . 

 MgO. 

 Kfo . 

 NaaO 



Per cent. 



46.14 

 17.07 



8.45 



11.70 



15.01 



.10 



1.11 



99.58 



The rock belongs evidently to the group of pyroxenites as described 

 by Williams,* but can not be classed under any of the varietal names 



* Am. Geologist, July, 1890, pp. 40-49. 

 Proc. N. M. 94 42 



