138 
James L. Gooch and Jeffrey S. Wiseman 
Table 2. Linear regression of eye and appendage lengths against body length. 
Values are for 25 males (upper row) and 25 females (lower row) per 
locality, averaged over 9 localities. Further explanation in text. 
Dependent variable 
slopes Y-interceptc 
Eye length 
0.04 
± 
0.02 
0.10 
.67 
0.01 
± 
0.02 
0.27 
.12 
Antenna 1 length 
0.62 
± 
0.16 
-0.41 
.69 
0.57 
± 
0.17 
-0.17 
.61 
Pereopod 7 length 
0.41 
± 
0.12 
-0.42 
.63 
0.50 
± 
0.09 
-0.33 
.80 
Uropod 3 length 
0.23 
± 
0.07 
-0.28 
.62 
0.20 
± 
0.06 
-0.20 
.65 
ficult. Holsinger and Culver obtained values of b close to unity and con- 
cluded that allometry was not important in the adult growth of antenna 
1, pereopod 7, or uropod 3 relative to body length. We also tested for 
allometry by solving for b in the equation Y = a b log X (Frazzetta 
1975), obtaining the slope b from plots of sample means of eye and ap- 
pendages against mean body lengths of males and females separately. 
Values of b ranged between 0.78 and 1.15 for appendage growth, none of 
which was significantly different from unity. For eye length, b for males 
was 1.08 ± 0.21 and for females 0.67 ± 0.28. The latter figure suggests 
negative allometry, but the large standard error and the anomalously low 
coefficient of determination for females (0.12) make this estimate of b 
meaningless. With qualification for eye length in females, we conclude 
with Holsinger and Culver that over the range of measurements used in 
adult G. minus allometry is a minor factor in determining length ratios. 
Interdemic regression data on sample mean eye and appendage 
lengths against mean body lengths are presented in Table 2. There are no 
significant differences in slope between sexes. Excepting female eye 
length, the coefficients of determination are between 0.61 and 0.80, in- 
dicating moderate scatter of locality means and probably reflecting non- 
uniform slopes and intercepts of growth equations of different popula- 
tions. The slopes for appendage growth are generally slightly higher than 
those obtained by Holsinger and Culver for Form III amphipods, but not 
significantly so. 
The ratios of eye and appendage length to body length in rank or- 
dered habitats are given in Table 3. Mean length of males is 9.28 ±0.16 
mm, about 1 mm less than Holsinger and Culver determined for III 
habitats. Mean female length is 6.73 ± 0. 14 or 73% of male length. There 
is considerable interdemic variation in female/male length ratio, from 
