METAMOEPHIO KOCKS. 273 



regard is paid to the occurrence in the field, on the other hand, the number 

 of possible hypotheses is generally reduced to one. 



There can be no question as to the regional character of the occurrence 

 of crystalline rocks at Knoxville. A part of the area of the map, to be sure, 

 is unaltered rock ; but from the westerly edge of the map westward the 

 crystalline rocks and serpentine form an unbroken mass many miles in 

 width; indeed, it would probably be possible to proceed from Knoxville to 

 the mouth of the Russian River, not in a perfectly straight line, but witli 

 no great deviations, without leaving this series. 



The crystalline rocks not eruptive. — It is equallv ccrtaiu that tliesc rocks are in 

 fact neither of igneous origin nor crystalline precipitates from an ancient 

 sea. No observer studying the rocks on the ground could fail to come to 

 this conclusion; and, if conviction be not brought home to the reader, it will 

 be due entirely to imperfect description. No area of more than a few yards 

 can be examined without revealing evidence that the rocks are stratified. 

 It is true that in a large proportion of cases there is entire discordance be- 

 tween the planes of stratification of different portions of a single cropping, 

 but fractures may often be detected between adjoining masses which bear 

 this relation, and sometimes distinct plication accompanied by a more or less 

 elaborately developed fissure system is apparent. In the granular and ser- 

 pentinoid series no masses are intercalated which exhibit the common char- 

 acteristics of eruptive rocks: a lack of stratification and a tolerably persistent 

 granular or porphyritic structure. The only rock in this district possessing 

 this character is the basalt, which is manifestly far younger than the strati- 

 fied rocks. It has frequently been maintained that certain rocks, like gneiss, 

 which show distinct stratification, are of eruptive origin. That a gneissoid 

 structure may bo produced by igneous action, at least over small areas, is 

 certain. I have myself seen such a case in New England. A dike of some- 

 what porphyritic diabase filled a fissure in unstratified granite, but at one 

 point an irregularity in the fissure left a mass of granite projecting into the 

 dike. This had been softened by the heat of the eruptive rock and molded 

 by the pressure of the intrusive material. It had assumed a perfectly 

 gneissoid structure without being separated from the wall of granular granite- 

 But when an igneous origin is attributed to large areas of rock it must at 

 MON xin — 18 



