July, 1955 
mercial fishermen who operate on the Mis- 
sissippi River. 
Wing nets with and without leads were 
used in the 1944 survey. In 1946, all 
wing nets were fished without leads. The 
meshes used in this study were 1 inch, 
114 inches, and 214 inches. The 114-inch- 
mesh wing nets were used only in 1944. 
The catch data for the 214-inch-mesh 
wing nets fished with leads have been 
combined with the catch data for these 
nets fished without leads. 
About 70 per cent of the commercial 
fish catch reported from the Illinois sec- 
tion and 59 per cent of that reported from 
the Missouri section of the Mississippi 
River for 1947 were taken in hoop nets 
(Upper Mississippi River Conservation 
Committee 1948, third section:19). The 
reports do not differentiate between hoop 
nets and wing nets; however, the wing 
net catch comprised only a small fraction 
of the total annual yield. 
The trap net is quite similar to the 
wing net in construction and operation, 
fig. 10. It differs from the wing net in 
that it has a double rectangular wooden 
frame in front of the first hoop. It is 
staked out in the same fashion as the wing 
net. It may be fished either with or with- 
out a lead. During the survey, trap nets 
were fished at only three stations. The 
mesh used was 114 inches. The nets were 
fished in localities identical to those fished 
with wing nets. 
Analyzing Entrapment Catch Data. 
—lIn the Mississippi River survey, entrap- 
ment devices were tested more than any 
of the other devices. 
The unit of measurement usually em- 
ployed in analyzing net catch data is the 
net-day, and the efficiency of a fishing de- 
vice is measured by the number of fish 
and/or the number of pounds of fish it 
catches per net-day. One net fished for 
24 hours is termed one net-day. In this 
investigation, the number of net-days 
fished (or trap-days in the case of basket 
traps) was recorded for each entrapment 
device. 
In some instances, when the catch per 
net-day of a certain species of fish in one 
type of net is compared with the catch of 
this same species in another type of net, 
the difference is obviously significant. In 
other instances, it may not be clear as to 
STARRETT & BARNICKOL: COMMERCIAL FISHING DEVICES 
Sey 
whether the catch difference is significant, 
unless the data are tested statistically. In 
this study, tests for significance of differ- 
ences in the efficiency of various entrap- 
ment devices were made by using the chi- 
square method of analysis (Snedecor 
1946:16). The chi-square was computed 
from the following formula: 
ge — Siam), (XK = my) 
my Me 
X, and X, are the actual catches in 
numbers of commercial-sized fish and m, 
and m, are the expected catches. The ex- 
pected catch for each net of the two nets 
being compared is that part of the actual 
total catch of the two nets which is di- 
rectly proportional to the total number of 
net-days fished by the net. In table 4 are 
given catch data on commercial-sized 
carp, as presented in table 12, and below 
the table an equation that tests whether 
there is a significant difference in catch 
of commercial-sized carp (15 inches or 
more total length) between 1-inch-mesh 
wing nets (without leads) and 114-inch- 
mesh wing nets (most without leads*) at 
the 0.05 level with 1 degree of freedom. 
The value 21.09 indicates statistically 
that at the 0.05 level the catch of com- 
mercial-sized carp is significantly greater 
in the 114-inch-mesh wing net than in the 
l-inch-mesh wing net. A chi-square value 
of 3.841 or more is considered significant 
at the 0.05 level with 1 degree of free- 
dom. The fish samples were taken at the 
same stations and the nets were fished si- 
multaneously, thereby eliminating station 
and seasonal differences with respect to 
species. However, at each station various 
habitats were netted, and as a result prob- 
ably different segments of the population 
at each station were sampled. This pat- 
tern of netting in no way affects the anal- 
ysis, since the information sought was, 
in many cases, for the purpose of giving a 
comparison of efficiency and selectivity of 
two nets designed for different habitats. 
For example, the hoop net usually is fished 
in deeper water than is the wing net and 
is always fished in the current; the wing 
net usually is fished in quiet water. Hab- 
itat differences of flowing and quiet water 
are discernible in the catch data, table 5. 
In the backwater areas and sloughs of the 
* Of 156.10 net-days. 8.84 net-days were with leads. 
