404 
Fishery Bulletin 99(3) 
Table 5 
Species which differed significantly in either mean numbers or total length (mm) from comparative samples collected from the 
commercial fishery and separated with or without the aid of a salt-box, and results of Wilcoxon signed rank tests used to compare 
means. 
Species 
Salt-box 
Mean +SD 
n 
No salt-box 
Mean ±SD 
n 
Z 
P 
Number of specimens 
bay anchovy 
2.7 ±4.9 
15 
0.3 ±0.6 
15 
3.159 
0.002 
Atlantic croaker 
28.8 ±14.5 
15 
6.9 ±5.0 
15 
4.030 
<0.001 
hardhead catfish 
2.5 ±3.3 
15 
0.7 ±1.4 
15 
2.150 
0.032 
bay whiff 
0.1 ±0.3 
15 
1.3 ±1.4 
15 
-3.097 
0.002 
bluntnose jack 
0 
15 
1.3 ±1.7 
15 
3.440 
0.001 
Atlantic spadefish 
0 
15 
0.5 ±0.9 
15 
-2.366 
0.018 
inshore lizardfish 
0 
15 
0.4 ±0.8 
15 
-2.073 
0.038 
brown shrimp 
2.0 ±1.8 
15 
1.1 ±1.8 
15 
3.397 
0.001 
Total length 
Atlantic croaker 
120 ±15 
419 
104 ±19 
93 
8.129 
<0.001 
gulf menhaden 
110 ±47 
120 
119 ±22 
57 
3.460 
<0.001 
spot 
98 ±23 
157 
109 ±20 
60 
3.910 
<0.001 
sand seatrout 
103 ±30 
45 
129 ±25 
44 
4.830 
<0.001 
pinfish 
75 ±13 
9 
89 ±20 
82 
-3.348 
0.001 
blue crab 
82 ±23 
60 
66 ±28 
16 
-2.268 
0.023 
Table 6 
Means of bycatch (all species) and Atlantic croaker initial percent survival, final percent survival, and mortality rate (M) and 
results of Wilcoxon signed rank tests comparing the means. All samples were collected from Texas bay shrimpers and separated 
with or without the aid of a salt-box. 
Variable 
Salt-box 
Mean ±SD 
n 
No salt-box 
Mean ±SD 
n 
Z 
P 
Bycatch 
initial survival 
76.1 ±21.7 
15 
55.5 ±35.1 
15 
1.763 
0.078 
final survival 
12.9 ±6.0 
15 
34.5 ±28.9 
15 
-1.389 
0.093 
M 
0.08 ±0.03 
15 
0.48 ±1.23 
15 
2.605 
0.009 
Atlantic croaker 
initial survival 
74.4 ±25.7 
13 
81.8 ±25.1 
6 
0.746 
0.046 
final survival 
28.6 ±19.9 
13 
62.3 ±30.7 
6 
2.330 
0.020 
M 
0.23 ±0.12 
12 
0.26 ±0.19 
4 
0.364 
0.716 
of M were similar between separation techniques. Means 
of initial (81.8% [±25. 1%] ) and final (62.3% | ±30.7%] ) sur- 
vival of fish separated without a salt-box were significant- 
ly greater than initial (74% [+25.7%]) and final (28.6% 
[±19.9%]) survival of those separated with a salt-box. 
By examining Spearman correlation coefficients that 
were significantly greater than 0, we found a relationship 
between the three measurements and factors other than 
use or non-use of a salt-box (Table 7). Initial and final 
survival estimates were negatively correlated to trawling 
time and catch separation time, whereas M was positive- 
ly correlated to the two variables. Only estimates of M 
(r s =0.488) were found to be related to the use or non-use 
of a salt-box. Final estimates of survival and M were also 
associated with some bycatch species. M was positively 
correlated to numbers of Atlantic croaker (r s =0.526) and 
sand seatrout (/' s =0.458) in the samples, whereas final sur- 
vival was inversely correlated to the two species (Atlantic 
croaker r s =-0.391, sand seatrout r s =-0.564). Other species 
to which M and final survival were associated occurred 
in fewer than half the samples or averaged <1 specimen 
per sample. Initial survival estimates were not related to 
