156 THE JOURNAL OF (GEOLOGY. 
To return to the analyses for a few moments, it will be seen 
on examination that the rocks of A‘gina are, on the whole, 
notably less acid than those of Methana; the main mass of 
/Egina (excluding such small eruptions as those of Anzeiou and 
Kakoperato) being made up of andesites with a silica content 
not exceeding 56 per cent., while a large part of Methana is 
composed of dacites with over 61 per cent., and the most basic 
rocks have not less than 59 per cent. or more. This difference is 
most strikingly shown on comparing such similar rocks as those 
of Mt. Oros and Mt. Chelona, they being both hypersthene- 
andesites of very similar mineralogical composition. Here we 
find that the silica percentage of the former is 55.87, while in 
the latter it is 59.83. A similar result is reached on comparing 
the analogous rock of Mt. Chondos with that of Kaimeni, the 
respective SiO, percentages being 55.46 and 59.94. It may also 
be noticed that the dacites of A‘gina are less acid than the 
dacites of Methana. When we look at these facts in the light 
of the law above laid down it follows that, on the whole, the 
eruptive activity of Methana most probably belongs to a later 
period than that of A®gina; which conclusion is supported by 
the fact that.the only known historical eruption of the region 
took place on Methana, as well as by the absence in the latter of 
andesitic breccia cemented by tertiary limestone, and by, the 
general appearance of the two islands. 
Iddings* has endeavored to show that, as a rule, ‘‘the order 
of eruptions for any complete series of igneous rocks is from 
intermediate to more and more extreme varieties” (Joc. cit. p. 
183), so that at the beginning rocks such as the andesites are 
poured forth while at the close the eruptions are of basalt and of 
rhyolite; the exceptions being explained as belonging to 
unfinished progress of differentiation and eruption. While the 
exceptions are numerous and often not explicable in this man- 
ner, yet, in many localities, this law does undoubtedly hold 
good, and it will be of interest to determine whether this law 
also applies in the region which is the subject of this paper, as 
t Orig. Ign. Rocks, pp. 144 ff. 
