ADDRESS. 6 



partly undergone metamorpliism. Under these circumstances, it is not 

 surprising that when chemically analysed, there is often little difference in 

 the constituents of the unmetamorphosed and the metamorphosed rock. 

 This is a point of some importance in relation to the origin and non- 

 primitive character of gneiss and other varieties of foliated strata, and 

 also of some quartzites and crystalline limestones. 



I am aware that in North America formations consisting of meta- 

 morphic rocks have been stated to exist of older date than the Laurentian 

 gneiss, and under any circumstances it is obvious that vast tracts of pre- 

 Laurentian land must have existed in all regions, by the degradation of 

 which, sediments were derived wherewith to provide materials for the de- 

 position of the originally unaltered Laurentian strata. In England, Wales, 

 and Scotland attempts have also been made to prove the presence of more 

 ancient formations, but I do not consider the data provided sufficient to 

 warrant any such conclusion. In the Highlands of Scotland, and in 

 some of the Western Isles, there are gneissic rocks of pre-Cambrian age, 

 which, since they were first described by Sir Roderick Murchison in the 

 North-west Highlands, have been, I think justly, considered to belong to 

 the Laurentian series, unconformably underlying Cambrian and Lower 

 Silurian rocks, and as yet there are no sufficient grounds for dissenting 

 from his conclusion that they form the oldest known rocks in the British. 

 Islands. 



It is unnecessary here to discuss the theory of the causes that produced 

 the metamorphism of stratified rocks, and it may be sufficient to say, that 

 under the influence of deep underground heat, aided by moisture, sand- 

 stones have been converted into quartzites, limestones have become 

 crystalline, and in shaley, slaty, and schistose rocks, under like circum- 

 stances, there is little or no development of new material, but rather, in 

 the main, a re-arrangement of constituents according to their chemical 

 affinities in rudely crystalline layers, which have very often been more or 

 less developed in pre-existing planes of bedding. The materials of the 

 whole are approximately the same as those of the unaltered rock, but 

 liave been re-arranged in layers, for example, of quartz, felspar, and mica, 

 or of hornblende, &c., while other minerals, such as schorl and garnets, 

 are of not infrequent occurrence. 



It has for years been an established fact that nearly the whole of the 

 mountain masses of the Highlands of Scotland (exclusive of the Laurentian, 

 Cambrian, and Old Red Sandstone formations), mostly consist of gneissic 

 rocks of many varieties, and of quartzites and a few bands of crystalline 

 limestone, which, from the north shore to the edge of the Old Red Sand- 

 stone, are repeated again and again in stratigraphical convolutions great 

 and small. Many large bosses, veins, and dykes of granite are asso- 

 ciated with these rocks, and, as already stated, it sometimes happens that 

 it is hard to draw a geological line between granite and gneiss and vice 

 versa. These rocks, once called Primary or Primitive, were first proved 

 by Sir Roderick Murchison to be of Lower Silurian age, thus revolu- 



