652 SCIENCE. 
October, 1883, about 2.30 p.m., the air became 
suddenly darkened, like night ; and soon after, 
a shower of mixed sand and water fell for 
about ten minutes, covering the ground with a 
thin layer. The windows were so coated that 
it was impossible to see through them. A 
small portion of the sand was referred to me 
for microscopical examination; and, at this 
time of general concern in atmospheric dusts, 
it may be well to note the products of one of 
the Alaskan craters, which for some time 
has been in a state of more or less vigorous 
activity. 
The sand is composed chiefly of crystalline 
fragments, of which felspar is the most abun- 
dant. It generally occurs in irregular, angular 
splinters, but not infrequently in well-preserved 
erystals with large inclusions and a distinct 
SERIES 1. 
on 
[Vou. IIL, No. 69. 
multitude of microlites. Very rarely particles 
of clear volcanic glass may be found, and they 
are generally freighted with grains of magne- 
tite. The grains of sand are fresh, and it is 
undoubtedly of recent volcanic origin. Its 
mineralogical composition is that of a horn- 
blende andesite. Mr. Chatard of the U. S. 
geological survey, who made a partial analy- 
sis of the sand, found 52.48 % of silica, — 
an amount which is certainly much below the 
average for hornblende andesite. This pau- 
city in silica, as well as in glassy particles, 
may be readily comprehended by considering 
the origin, composition, and distribution of 
volcanic sand and dust. 
Among the various sands and dusts exam- 
ined and compared for the purpose of seek- 
ing an explanation for the poverty of silica in 
"Swe yy ee eS 4 
zonal structure. These crystals are slightly 
tabular, parallel to the clinopinacoid, and, lying 
on that plane, they present an approximately 
hexagonal outline, about 0.15 of a millimetre in 
diameter. Lath-shaped felspars, so abundant 
in basalts, were not observed. A few thin 
cleavage lamellae, with parallel extinction, 
showed no banding due to polysynthetic twin- 
ning; but by far the greater portion of the 
fragments, in polarized light, were distinctly 
striated. The pale green augite appears usu- 
ally in the form of broken prisms; and the 
deep brown, strongly pleochroitic hornblende, 
which is less abundant than either the augite 
or felspar, occurs for the most part in cleav- 
age plates. Irregular grains and crystals of 
magnetite complete the list of minerals which 
form an essential part of the sand. Crystal 
fragments of the minerals already mentioned 
constitute the largest portion of the sand ; but, 
besides these simple grains, there are others 
complex in their nature. They correspond to 
the groundmass of a porphyritic rock, and are 
composed of an amorphous base, containing a 
—<— sf == > 
[>> = Fee ee 
the sand from Unalashka, is one collected about 
a dozen miles north-east of Mount Shasta, in 
northern California. ‘This sand is considerably 
coarser than that from Unalashka, and is com- 
posed chiefly of crystal fragments of felspar, 
augite, hypersthene, hornblende, and magne- 
tite, with particles of microlitic groundmass, 
and considerable pumiceous glass. The min- 
eralogical composition of the sand is the same 
as that of the hornblende andesite which issued 
from the prominent and well-preserved crater 
named, by Capt. Dutton, Shastina, upon the 
north-western flank of Mount Shasta. There 
can be no doubt that the sand was ejected 
from Shastina, for all the other craters of that 
region have erupted different material. Mount 
Shasta itself has effused hypersthene andesite, 
and the smaller craters to the eastward have 
furnished basalt. According to Mr. Chatard’s — 
analyses, the sand contains 60.92 % of silica, 
while the Shastina lava, to which it belongs, 
contains 64.10 % of silica. 
As far as definite observations | upon this 
subject have been made, it appears to be true, — 
