18,3 Smith: Tropical Geology and Engineering 235 
the more nearly exact work possible in the stratified rocks. The 
geologist can do many things, but he is not gifted with second 
sight. 
Dams and reservoirs. — The geological conditions affording the 
best sites for dams and reservoirs have also been discussed by 
several writers. With reference to this phase of the subject, 
the writer wishes to state that a geological examination of a 
district and several years of stream gauging are not only desir- 
able but imperative in the Tropics. Many examples of how 
these works should not be located could be cited, but two or 
three will suffice. The city reservoir at Montalban, Luzon, was 
located in a limestone gorge. The limestone is full of caverns 
and the reservoir failed to hold water until it was lined, bottom 
and sides, with cement. If the dam had been placed at the 
upper end of the gorge the reservoir would have had a natural 
clay bottom. 
The destruction some years ago of the Tarlac (Luzon) dam 
probably could not have been foreseen, but if a few more data 
concerning annual precipitation and stream discharge had been 
secured the trouble might have been averted. Agno River 
when under full head is a terrible engine of destruction. 
Another dam was built in the hills back of Cebu, making use 
of an apparently substantial formation for abutments and spill- 
way. Drilling was recommended by the geologist who examined 
•it; but no, that would be too expensive. The abutments held 
in a terrific storm that ensued; but the spillway, which was 
too small, forced the water to eat its way down through a de- 
composed formation which had a hard shell on the outside, so that 
the dam was rendered useless. The writer, from his observa- 
tions in the Philippines, should be inclined to say never when 
it can be avoided build a reservoir in the Tropics. If water 
is needed, use diversion weirs in the streams or artesian wells. 
Coal mining. — The writer had not intended to touch on any 
phase of the application of geology to mining engineering in 
this article for the reason that the connection between the two 
is so generally recognized. There is one particular phenomenon 
that has given more or less trouble to those who would work 
coal seams in the Far Eastern Tropics. It has been the writer’s 
experience throughout some years of examination of coal pros- 
pects and mines in the Tropics that the persistence of a given 
seam of coal is always a matter of conjecture. The seams are 
either interrupted by small faults or they thin, or “peter out,” 
just when you are counting on a good working thickness. 
