492 J. Allan Thomson—Inclusions in Volcanic Rocks. 
enclosing rocks, and here the xenoliths have probably a deeper 
significance.’’ For scattered crystals of foreign origin he uses the 
term sporadic xenocrysts. This ‘genetic relationship’ of cognate 
xenoliths to their host is not to be confused with that of homceogenous 
inclusions. They are to be understood as fragments of rocks already 
consolidated which belong to the same petrographical province. Thus 
the inclusions of gabbro in granite in Skye are considered cognate 
xenoliths. Lacroix regularly includes such cases amongst the enallo- 
genous inclusions. 
Judd, in describing the glomeroporphyritic structure of the Fair Head 
dolerite, says': ‘‘Those who regard the enclosures composed of olivine, 
enstatite, augite, etc., which are found in some basalts as having 
separated at some period from the magma, out of which the 
surrounding rock was itself formed, may regard these as examples of 
the glomeroporphyritic structure on a gigantic scale.’ Conversely, we 
may regard the small groups of anorthite and olivine crystals, with 
the structure of troctolite, which lie scattered through the dolerite, as 
homceogenous inclusions of small dimensions. 
Throughout this paper, then, I shall speak of inclusions as 
enallogenous or homceogenous, dividing the former into xenocrysts 
and cognate or accidental xenoliths. 
The study of xenoliths is important because of the alterations that 
they have undergone under the influence of the enclosing magma. 
In other words, they furnish the best) materials for the study of the 
contact metamorphism, both exogenous and endogenous, of volcanic 
rocks.? It is rare to find strongly marked contact phenomena at the 
junction of volcanic rocks with older rocks, but even where this is 
the case, as with the Portrush rock described below, the inclusions 
generally show the same phenomena more strongly marked and more 
easily studied. Lacroix has found from the study of these inclusions 
that a broad division of volcanic rocks must be made into the two 
classes trachytoid and basaltoid. The former contain orthoclase or 
acid triclinic felspars, the latter are without these minerals. This 
division is set up because in the former the modifications undergone 
by the inclusions are often intense, are of a chemical order, and can 
only be explained by the presence of mneralisators, while in the 
latter the metamorphic phenomena are less intense and could be 
produced by the action of heat alone, or, if of a chemical order, are 
limited to the surface of the inclusions. 
The absence of mineralisators in basaltoid lavas is to be explained 
by their more complete fluidity, permitting of the escape of the gases. 
That mineralisators are present with the magmas in depth has been 
shown by Lacroix in his study of the ophites and lherzolites of the 
Pyrenees.’ The trachytoid lavas, on the other hand, being much more 
1 Q.J.G.S., vol. xlii (1886), p. 71. 
* Lacroix has summed up his conclusions on this subject in a paper ‘‘ H’tude sur 
le Métamorphisme de Contact des Roches Volcaniques’’?: Mémoires présentés par 
divers savants a |’ Académie des Sciences, tome xxxi (1894), No. 7. 
3 See A. Lacroix, ‘‘ Les Phénoménes de contact de la Lherzolite et de quelques 
Ophites des Pyrénées”’: Bull. Carte Géologique de la France, No. 42 (tome vi, 
1895), ch. iii, p. 124. 
