236 N. L. BOWEN 



pyroxenites the case is apparently the same. Limburgite and 

 augitite can scarcely be regarded as monomineralic rocks in the 

 stricter sense of the term, and that is the only sense in which it can 

 be used in testing the hypothesis. The presence of both pyroxene 

 and olivine, of a glassy base and usually of some feldspathoid makes 

 it clear that these effusive pyroxenites do not constitute an exception 

 to the rule that the monomineralic rocks do not have effusive 

 equivalents. Apparently, the facts are in accord, therefore, with 

 the hypothesis that monomineralic rocks are accumulated masses 

 of crystals. 1 Mention may be made again here of Vogt's discovery 

 that the richer a peridotite is in olivine, the richer the olivine is in 

 magnesia, a fact which is readily explained on the assumption that 

 peridotites are made up of accumulated early crystals. 



All of the monomineralic rocks often do occur, however, in a 

 manner which has led a very great number of investigators to 

 speak of the magmas of these rocks as freely as of the magmas of 

 any others. This is probably due partly to the fact that the possi- 

 bility of their origin after the manner here advanced did not occur 

 to the investigators, but whether this was always the case is a 

 question that, again, an individual cannot answer. One of the 

 most remarkable occurrences of anchi-monomineralic rocks, espe- 

 cially pertinent in the present connection, is that described by 

 Harker from the islands west of Scotland. As a result of his 

 minute descriptions an especially favorable opportunity is offered 

 of discussing these rocks in the light of the present conception of 

 the origin of monomineralic rocks. The rocks are intricately 

 banded in such a manner as to lead Harker to suggest the intrusion 

 of a non-uniform magma, implying apparently a non-uniform 

 liquid. 2 A difficulty in the way of accepting this interpretation is 

 that connected with obtaining a non-uniform liquid, especially with 



1 While it has been necessary in applying the foregoing tests to set aside anchi- 

 monomineralic rocks containing a considerable amount of other minerals, it should 

 not be assumed that there is any essential difference in the method of formation. 

 These are merely examples in which accumulation of crystals of one kind has not taken 

 place to quite the same degree and which consequently could have had a consider- 

 able amount of interstitial liquid. 



2 "Geology of the Small Isles of Inverness-shire," Mem. Geol. Survey Scotland, 

 1908, p. 74. 



