14 NEW YORK STATE MUSEUM 



able character that their relations can not be stated definitely. 

 In St Lawrence county the sedimentary gneisses are widespread 

 in the vicinity of tlje limestone belts above mentioned. They have 

 also been traced by the writer to the east toward the interior as 

 far as Cranberry lake. They are the country rock of the magnetite 

 deposits in this section. Professor Kemp has described gneisses 

 of sedimentary type in southern Essex, Warren and Washington 

 counties. As to the southern border of the Adirondack region, 

 little has been made known but it is probable that the sedimentary 

 gneisses are well represented. 



Amphibolite. Involved with the limestones and gneisses, and 

 less frequently with the plutonic igneous rocks, are small masses 

 of amphibolite, dark colored and consisting essentially of horn- 

 blende and feldspar. They often have a rusty appearance that 

 betrays the presence of pyrite. Their occurrence in tabular bands, 

 which may be persistent for considerable distances, is suggestive 

 of dikes and it is quite likely that they are in part metamorphosed 

 diabases or gabbros. This view is particularly applicable to 

 examples that have a plagioclase as the principal feldspar constitu- 

 ent, but can hardly be accepted for occurrences in which the horn- 

 blende is associated with orthoclase, as is not infrequently the 

 case. For these the derivation from a magnesian shale seems to 

 be the more obvious explanation. 



Quartzite. As a somewhat uncommon type of the Precambric 

 sediments, may be mentioned the occurrence of quartzite which 

 has been made known on both the eastern and western borders of 

 the Adirondacks. 



In Essex county, Professor Kemp has noted several localities 

 where this undoubted fragmental rock occurs. It nearly always 

 carries graphite, pyrite and sillimanite and sometimes feldspar 

 and mica. At Hague on Lake George and at the village of Graph- 

 ite, 5 miles west from Hague, a bed up to 15 feet thick is included 

 between a garnetiferous sillimanite gneiss. At Rock pond between 

 Graphite and Hammondville, there is another area; while in the 

 town of Lewis, 3 miles south of Elizabethtown, exposures show a 

 thickness of 100 feet of quartzite overlain by graphitic gneiss. 



Professor Smyth has found the same rock in St Lawrence county. 

 On Wells island in the St Lawrence river a white vitreous quartzite 

 is exposed along a ridge for nearly 5 miles with an estimated thick- 

 ness of 500 feet. It is associated with schist and both are cut out 

 by granite gneiss which forms the southern part of the island. 

 A second belt occurs between Redwood and Rossie, the quartzite 



