SILICA. 271 



About a mile and a half north of Lewis corners, there is another locality. The mineral 

 here is snow-white, and closely resembles that from Easton in Pennsylvania. It is associated 

 with garnet and quartz, the latter of which gives a sharpness to the masses of tabular spar, 

 which it does not usually possess. 



At Rogers' rock, near the line between this and the county of Warren, tabular spar occurs 

 with characters similar to those observed in the specimens from Willsborough. It is asso- 

 ciated with garnet, adularia and common feldspar. 



The same mineral is not unfrequently found in loose masses in various parts of this county. 

 It has thus been met with in the vicinity of Port Ileury. As thus found, it has a greyish 

 white colour, with a tinge of olive, and is highly translucent. It is associated with brown 

 mica, and sparingly with scales of graphite.* 



Lewis County. In the town of Diana, about a mile from the Natural bridge, is a large 

 deposit of tabular spar in the limestone. It is of a snow-white colour, has a vitreous lustre, 

 and closely resembles some varieties of tremolite. Its apparent purity induced mc to subject 

 it to analysis. Its composition is as follows : 



Silica, 51 . 90 



Lime, 47.55 



O.xide of iron, 0. 



o.^ 



:j 



At this locality, the tabular spar is associated with green coccolite, and is exactly similar 

 to the specimens which are found in loose masses in Oneida county. I have no doubt that 

 this is the rock from which these bowlders have been detached. 



There is another locality nearer to the Natural bridge than the preceding, where the mineral 

 is abundantly found in detached masses. Here, however, it is of a greyish colour, and the 

 folia are not so large or beautiful, nor is the lustre so high. 



Oneida County. Large bowlders of tabular spar, associated with garnet, have been dug 

 up in the village of Boonville. The specimens, as I have just stated, resemble those from 

 Lewis county more nearly than those from Essex. The Essex county specimens are usually 

 associated with garnet or colophonite, whereas those from Boonville and Diana have green 

 granular pyroxene (coccolite) as their associate. I have compared specimens from the two 

 localities, and find it very difficult to distinguish them. This is a fact of some interest in 

 connection with the mode in which bowlders have been transported. Boonville is between 

 forty and fifty miles directly south of the Natural bridge. These bowlders have been described 

 by Prof. Hubbard, of Dartmouth College. t 



• Shepard. American JouTnal of Science. XVH. 110. ] American Journal of Science. XXXII. 230. 



