30 
in abundance. Reaching Stevens Creek during a freshet, we found 
much drift of floatage, which was not the case in the main body at 
Parr Shoals Pond, where these hammocks were found. The larvie on 
a hammock were frequently of different sizes and were sometimes in 
places to which they could not have drifted, i. e., inside of the ham¬ 
mock, and were, it appeared, from eggs deposited there. It is worth 
noting that minnows were very common around these hammocks. 
We found larva) in them occasionally in water probably 20 to 25 feet 
deep. 
MOVEMENT OF LARV7E ON THE PONDS. 
The freshet at Stevens Creek put in motion a great deal of drift 
and floatage, which went down stream, and then when the Savannah 
River, into which the creek empties, rose the floatage stopped and 
then drifted with the winds. At first the fine stuff and small trash was 
much scattered, but it was collected together within a few days, and in 
a few more went to the banks or occasionally lodged against obstruc¬ 
tions in the stream. This floatage carried Anopheles larvae in large 
numbers. They drifted about across the pond and up and down. At 
first these were of all sizes, but practically no pupae; by the end of a 
week the very small ones had disappeared, the larvae were nearly all 
full size and pupae were present, the latter being more generally found 
in a resting place than adrift. Full-grown larvae do not merely drift 
about on floating objects; they possess considerable power of selection 
as to where they will go in open water. Griffitts put a full-grown An¬ 
opheles larva in the water in Hellers Creek Bayou and followed in a 
canoe some yards off. It went to a piece of bark. This he submerged 
slowly, and the larva went off in a fairly straight line for, say, 20 
feet and attached itself to a leaf. It left this voluntarily and went 
about 15 feet at an angle to its former course to another leaf. This 
was sunk, and so several times. The larva went within 20 minutes 
at least 90 feet from its starting point over a course of approximately 
120 feet. Frequently we would sink a dipper of floatage containing 
larvae down under the water to get the larvae separate. They would 
swim off — for some feet or yards even—and attach themselves to some 
object floating on the water. Sometimes they showed distinct prefer¬ 
ence, passing by bits of drift to attach themselves to other pieces, 
or letting go one piece to select another. They are bv no means, 
then, entirely dependent on the drift of the object to which they are 
attached-for change of place. When there is any motion on the 
water, however, we have never seen them leave the object to which 
they were attached unless frightened. 
Over a considerable part of Stevens Creek Pond the breeding was 
profuse—the heaviest general production we have ever seen in a large 
