SKETCH OF 4GINA AND METHANA. 45 
search failed to diclose more than a very few which gave paral- 
lel or approximately parallel extinction. Moreover, the chemical 
composition of the rock is that of a typical dacite, and not at 
all that of a quartz-bearing trachyte or rhyolite.t ~The percent- 
age of K,O (1.66) is much too low for a rhyolite as rich in 
orthoclase as Lepsius makes out the rock to be. The percent- 
age of CaO (2.98) again is much higher than that shown by any 
other rhyolite analysis, while quite in accordance with other dacite 
analyses. It is true that itis the lowest lime percentage of any of 
the rocks of the district, but this is quite in accordance with the 
comparative scarcity of feldspar, and its position in the albite- 
anorthite series. For these reasons the feldspar is held to be a 
plagioclase, and the rock is consequently a dacite.’ 
Tuff—Only two specimens of this fragmental rock were 
collected. That from the west end of Mt. Chondos is light gray 
in color, rather finely granular, very friable, and shows evidence 
of stratification. Under the microscope it is seen to be composed 
of fragments of fresh olive-green hornblende, plagioclase crystals, 
magnetite grains, a few colorles augite crystals, and much micro- 
litic groundmass with colorless glass base, patches and streaks 
of brown dusty material being very abundant, and there being no 
definite order in the arrangement of the above fragmentary con- 
stituents. No remains of organic beings, either animal or vege- 
table, were to be seen, and the specimen is evidently a tuff of 
the light gray hornblende-andesite deposited subaérially. 
The brown compact tuff from farther east, near the top of the 
Chondos ridge, offers more points of interest. Megascopically 
it is very compact and fine grained, light brown and with a sub- 
greasy luster. In this main mass are many rounded enclosures 
of a gray porphyritic hornblende-dacite, showing hornblende, 
plagioclase, and quartz phenocrysts, and resembling the dacite of 
Kakoperato. 
Under the microscope the tuff shows a brecciated structure 
tCf. Analysis 15, in Part ITI. 
2ZIRKEL (Op. cit. II. p. 258), on the authority of Lepsius, classes these rocks, as 
well as the hornblende-andesite of Poros, with the rhyolites. 
