EXPERIMENTS ON LYNN MARSHES, MASS. 21 
covered with tanglefoot and were placed in the direction toward 
which the wind was blowing, the idea being that if the caterpillars 
were blown from the box which was set up at the central station, speci- 
mens might be caught on the screens. Owing to the variability of the 
air currents and the continued changes in the direction of the wind 
it was found better to use three screens, which were set up at distances 
ranging from 50 to several hundred feet from each other, but all at 
an equal distance from a central station. On May 6 a single screen 
was set up 50 feet from the central station. The wind was light and 
variable, but 10 minutes after it had been placed in position a cater- 
pillar was caught. At the expiration of an hour five caterpillars 
were found on the screen. 
It was then set up 100 feet from the central station. Four cater- 
pillars were caught in a short time. On the following day two extra 
screens were made in order to provide for changes in the direction of 
the wind. During the day two caterpillars were caught at 150 feet, 
two at 200 feet, one at 250 feet, and one at 300 feet. On May 10 one 
caterpillar was caught at a distance of 850 feet from the central 
station, after the screen had been exposed for 3-| hours. On May 11 
one caterpillar was caught 500 feet from the central station, after 
this screen had been exposed for about an hour. The screens were 
then put up GOO feet from the station, and 30 minutes later two 
caterpillars had been caught. Two of the screens were then set up 
TOO feet from the station and remained there until late in the after- 
noon. Before leaving for the night the remaining screen was set up 
in the direction of the prevailing wind and at a distance, which later 
proved to be 1,833 feet, from the central station. On the following 
morning a caterpillar was found on the last-mentioned screen. The 
supply of first-stage caterpillars being practically exhausted on this 
date, we were obliged to discontinue these experiments, but the screens 
were removed to another point on the marsh and set up at measured 
distances of one-fourth mile. 1.800 feet, 2,300 feet, and 2,800 feet 
north of Oak Island, Kevere, Mass. This is a wooded island (PI. X) 
of several acres surrounded by salt marsh; the trees are generally 
infested with the gipsy moth, and owing to its proximity to the ocean 
the growth of the larvae is considerably retarded. It was hoped that 
evidence might be secured that caterpillars were carried by the wind 
from this locality, but no larvae were caught on the screens, although 
they remained in position until June 13. 
in connection with these experiments it should be said that prac- 
tically the whole of this salt-marsh area (PL IX) is flooded at high 
tide ; hence it was possible to carry on the experiments during only a 
part of the day. The weather during April was cool, which is usually 
the case in this region near the seacoast, where the land is flat and 
exposed to strong air currents. The gipsy-moth caterpillars are not 
