SKETCH OF GINA AND METHANA. 153 
lime and magnesia, with alumina and the iron oxides quite vari- 
able. But a specially interesting point is the comparison of 
their analyses with those of their surrounding rocks of which 
three examples are given. The most striking fact is that in all 
three the segregations are more basic than their hosts, in one 
case the former containing 8.19 per cent. of SiO, less than the 
hornblende-dacite which carries it. They are also notably richer 
in lime, the same case as above containing 2.3 times as much as 
its host. They are also richer in magnesia. In general they are 
poorer in alkalies, though this not so prominent, but there 
seems to be no fixed relation as regards the alumina or the 
iron oxides. These results, which are what we would have been 
led to expect from the microscopical examination and our pre- 
vious knowledge of segregations', confirm us in the belief that 
we have in these masses to do with true segregations (endoge- 
nous enclosures) and not (exogenous) enclosures. 
Mention has already been made of the fact that the rocks of 
the Stavro district, Mt. Oros, Mt. Kouragio and Mt. Chelona are 
extremely poor in, if not quite free from, segregations, while on 
the contrary the dacites (with the exception of the Kolantziki 
rock and the Poros and Kaimeni andesites are rich in them. 
Now on referring to the analyses it will be seen that the rocks 
poor in segregations belong to the more basic members, while 
those carrying many of them are quite or very acid. The rocks 
of the Monastery District seem to occupy an anomalous position 
in this respect, since they generally abound in segregations, while 
at the same time the only one analyzed (No. 3 from Mt. Chon- 
dos) contains only 55.46 per cent. of silica. It must be noted, 
however, that the Mt. Chondos rocks carry fewer of these 
masses than the rocks of Mts. Gaiapha and Dendros, and it is 
possible that analysis of these last will show them to be richer 
in silica than No. 3. 
Leaving these rocks then out of the question, the poverty of the 
basic rocks in the basic segregations points, like the consideration 
above brought forward, to the conclusion that these masses are 
™Cf. ZIRKEL, op. cit, II., p. 788. 
