THE CLASSIFICATION OF IGNEOUS ROCKS. 471 



classification has at the present day no more than a historic 

 interest ; perhaps the best expositions of it are contained 

 in the writings of some of the Scottish geologists, Jameson, 

 MacCulloch, etc. 



There are, however, certain relics of Wernerism still 

 extant, two of which may be mentioned in passing. One 

 is the theory, still held in various forms, that most of the 

 crystalline schists and gneisses have been formed at a period 

 of remote geological antiquity, and no similar rocks have 

 been produced since the date of the "Urgneiss" or 

 4i Grundgebirge ". The other is the doctrine of an essen- 

 tial distinction between the " older " and " younger " igneous 

 rocks, on which some remarks have been made in an earlier 

 paper in this journal. 1 



Apart from these ideas, which appear somewhat crude 

 and arbitrary in the light of modern researches, various 

 attempts have been made to incorporate the chronological 

 principle in one form or another in classifications of igneous 

 rocks based on wider generalisations. It has long been 

 recognised that eruptions, whether deep-seated or super- 

 ficial, have not been equally distributed over the geological 

 time-scale. It is indeed the long pause in igneous activity 

 in Mesozoic times that has lent colour to the doctrine of an 

 older and a younger series of rocks already alluded to, and 

 closer examination reveals the existence of at least three or 

 four maxima of activity during the ages contemplated by 

 stratigraphical geology. The correspondence, as regards 

 both time and space, of igneous activity with the great 

 crust-movements that have shaped the continental masses, 

 and determined the dominant mountain ranges, is a subject 

 upon which we cannot enlarge here, but some have thought 

 to find in this a philosophical starting-point for a genetic 

 grouping of igneous rocks. The recurrence of identical 

 types at widely separated epochs, however, forbids the idea 

 of a systematic classification on such a basis. If, to fix our 

 conception, we imagine with Bertrand (24) that all the 

 rocks belonging to any one period of crust-movement and 



1 "Science Progress," vol. ii., pp. 48-63, 1894. 



