482 
THE VOLCANOES OF GUATEMALA. 
were deposited as ejecta from the old crater on the summit, and not 
from one in the position of the new crater. The thick beds of ash 
with their deep barrancos extended far beyond the left of the plate. 
All this presents a striking resemblance to the corresponding localities 
on the slopes of the Soufriere. On some of the ridges in the lower 
ground, as, for instance, in the coffee plantations, the resemblance 
is still more striking. Plate YI. is practically indistinguishable from 
a plate to appear in the Soufriere Report, Part II., and may be com¬ 
pared with Part I., Plate 35, which shows a similar place directly 
after the eruption. In each the ridge and slopes had been covered with 
a thin layer of ash. On the ridge this only received the rain which 
actually fell on it. This mostly sank in, and a firm crust was produced 
which offered considerable resistance to further change. On the slopes 
on each side the ash was exposed not only to the rain which fell directly 
on it, but also to the wash from the higher parts, and in many places 
had been carried away, and thus exposed the soil, on which vegetation is 
returning in many cases from the old roots. 
The loss of life is supposed to have been very great, but, un¬ 
fortunately, no accurate statistics are available, as the victims were 
chiefly Indians who had come down from their villages in the mountains 
into the coffee zone to assist at the harvest. Still the opinion of those 
on the spot puts it at possibly two or three thousand. 
The lake of Atitlan is not only one of the most interesting, but also 
one of the most beautiful places in the world, and its interest is much 
increased by the survival of several villages of Indians who retain many 
of their primitive customs, and still wear curious costumes. The lake is, 
roughly speaking, nearly circular, or would be so if it were not for several 
big volcanoes on its south bank, beyond which the plateau slopes rapidly 
down to the coastal plain. Its longest diameter is about 20 miles. On 
the east, north, and west, where there are no volcanoes, the sloj3es are 
usually very steep, though in a sufficiently advanced state of denudation 
to be a good deal cut up by valleys of rivers and brooks which flow into 
the lake. It has generally been supposed that the basin of this is only 
a continuation and union of these valleys, and that after they had been 
excavated, the volcanoes broke out on their beds and formed the lakes by 
blocking the exit for the water. This supposition is certainly plausible. 
The north shore is formed of volcanic tuffs and conglomerates of recent 
geological age, and sufficiently denuded to agree with either this or 
with the hypothesis that the lake itself is a crater, while the dip of 
their beds is so complicated that its evidence is not conclusive either 
way. I noticed, however, that the west shore of the lake extended in 
a well-marked, almost precipitous bank right round to the south of the 
volcanoes of San Pedro and Atitlan, and was perfectly separate from the 
slopes of the former, and to a large extent from those of the latter. I 
visited it, and found it composed of beds of tuff all dipping to the south 
