364 
“Naturalist’s” type (see pages 367, 368), supplemented in some situa- 
tions where there was unusually soft mud and where the heavier framed 
iron dredges were inclined to sink too deeply and fill too quickly, with a 
lighter framed dredge following closely a recent design by Ekman which 
was intended for quite a different purpose. (See Fig. 3 and 4.) 
Although there was no expectation early in the work of making 
more than a very rough quantitative application of the biological data 
obtained, all the dredge hauls were, from the first, of a previously de- 
termined and recorded length. The introduction into use in the summer 
of 1914, for work in water under eighteen feet in depth, of the “mud- 
dipper” (see Fig. 5), an instrument bearing some resemblance to 
the Walker dipper-dredge as used by Baker (1916, 1918), and the adop- 
tion of finer meshed inner bags for it and the dredges, was the means of 
what appeared to be rather more accurate work that year than in the 
first season, while at the same time its use in parallel test hauls of differ- 
ent lengths alongside the iron dredges suggested that averaged results 
from measured drags, under certain limits of length, with either, had a 
greater quantitative value than we had at first believed. It was found, 
in brief, that with a 22” & 6” front iron dredge we took on the average 
as many bottom animals by hauling five feet as by hauling ten, and with 
a 6-inch mud-dipper as many in two feet as in four, but that 
in hauls under two or five feet, in either case, we got less. As the aver- 
age 5-foot haul with the dredge was in the neighborhood of ten times the 
2-foot drag of the dipper, and the 2-foot dipper haul about five times a 
quick deep dip of the mud-dipper to a depth of about three inches (ap- 
proximate area covered, 25 square inches), it was an easy step to the con- 
clusion that on a rough average, if a few apparently aberrant cases be 
excluded, the most of the 5- to 10-foot dredge hauls might be safely 
taken to represent an effective drag of about one square yard, and the 
2- to 4-foot hauls of the dipper an effective drag of about 0.1 square 
yard (125 to 130 square inches). Still more recent parallel tests of the 
dipper alongside a new Petersen self-closing bottom sampler have not 
served materially to change these conclusions. 
The method used for collecting the small weed animals in the zones 
of densest vegetation (usually in water under four feet deep) was in- 
complete, taking in only the small fauna within the 0 to 9-inch depth line. 
A large bucket of known depth and diameter was lowered about the tops 
of the plants, the stems were cut off underneath, and then the bucket 
was brought into an upright position quickly ; after which the weed-tops 
were shaken out in the water saved, and that was finally passed through 
a 120-mesh sieve. Pulling up the weeds entire in water over two and a 
half feet deep had shown that the attached weed animals, whether snails, 
insect larvae, or Crustacea, were, in bulk at least, decidedly most abun- 
dant nearer the top. And the adoption of the method also followed, by 
necessity, some unsatisfactory experience in the use of a small 3-legged 
caisson and pump—which involved the handling of vastly more material 
