Graham et al.: Prosomal-width-to-weight relationships for L/mulus polyphemus 
241 
Table 2 
General linear model F-values and P-values for horseshoe crabs ( Limulus polyphemus) that were collected during the Horseshoe 
Crab Research Center trawl survey which sampled inshore continental shelf waters between New York and Virginia. Values are 
listed for all horseshoe crabs combined, immature and mature females, immature and mature males, mature females and males, 
and immature females and males. The prosomal-widthl PW )-to-weight relationship was analyzed for various combinations of sex 
and maturity stage (Mat). Significant interactions, after Bonferroni adjustment, are indicated by an asterisk. 
Variable 
All data 
(df=7, 924; n=925) 
Mature females 
vs. 
Immature females 
(df=3, 481 ;ti = 482) 
Mature males Immature females 
vs. vs. 
Immature males Immature males 
(df=3, 442; w =443 )(df= 3, 346; tj = 347) 
Mature females 
vs. 
Mature males 
(df=3, 577; 72 = 578) 
F 
P 
F 
P 
F 
P 
F 
P 
F 
P 
PW 
3770.45 
<0.0001* 
2490.51 
< 0.0001* 
1456.77 
<0.0001* 2849.41 
<0.0001* 
2056.93 
<0.0001* 
Sex 
0.00 
0.9866 
— 
— 
— 
— 
6.17 
0.0135* 
3.46 
0.0635 
Mat 
0.20 
0.6582 
1.46 
0.2271 
5.61 
0.0183 
— 
— 
— 
— 
PW x Sex 
1.16 
0.2815 
- 
— 
— 
— 
6.76 
0.0097* 
2.19 
0.1396 
PWxMat 
1.59 
0.2075 
0.98 
0.3234 
5.51 
0.0194 
— 
— 
— 
— 
Sex x Mat 
6.86 
0.0090* 
— 
— 
— 
— 
— 
— 
— 
— 
PW* Sex* Mat 
6.18 
0.0131 
— 
— 
— 
— 
— 
— 
— 
— 
Table 3 
The number of individuals sampled (n), coefficient values ( a, b ), standard errors for coefficients (SE [a], SE [6] ), and correlation coef- 
ficient (r 2 ) of the relationship between prosomal width and weight for horseshoe crabs (Limulus polyphemus ), log e ( Wif)=log ( ,(PVTx 
o +log e (fi). Samples were collected during the Horseshoe Crab Research Center trawl survey (i.e., inshore continental shelf waters 
between New York and Virginia), spawning surveys (i.e.. New Jersey, Delaware, New Hampshire), and the commercial fishery 
(i.e., Delaware). All regressions are significant. 
n 
a 
b 
SE(a) 
SE(b) 
r 2 
Female (all) 
1025 
2.98 
-15.71 
0.02 
0.10 
0.96 
Females (mature) 
802 
2.65 
-13.85 
0.04 
0.21 
0.86 
Females (immature) 
223 
2.85 
-15.10 
0.05 
0.23 
0.95 
Males (all) 
1055 
2.89 
-15.39 
0.02 
0.12 
0.94 
Males (mature) 
931 
2.97 
-15.80 
0.02 
0.13 
0.94 
Males (immature) 
124 
2.58 
-13.81 
0.10 
0.50 
0.85 
population average about 80% of the prosomal width 
of the females (Shuster, 1979) and mature females are 
significantly heavier than mature males because of their 
larger size and added weight associated with numer- 
ous eggs within their prosomas (Leschen et al., 2006). 
Therefore, it is inappropriate to use the same conversion 
factor for both sexes. 
Conversion factors should also vary by state, to take 
into account the larger size and greater weight of horse- 
shoe crabs in Mid-Atlantic states. Horseshoe crabs from 
the middle Atlantic region are significantly larger than 
animals from Cape Cod Bay to Maine and those from 
the Gulf of Mexico (Shuster, 1979). Morphometries 
(Shuster, 1979; Riska, 1981), survey data on the dis- 
tribution of horseshoe crabs along the continental shelf 
(Botton and Ropes, 1987), and population genetic stud- 
ies (King et al., 2005) strongly indicate that there are 
geographically distinct breeding populations throughout 
the range. Some intermingling of populations occurs 
along the middle Atlantic coast, especially from New 
Jersey to Virginia (Swan, 2005), where much of the 
trawl-based fishery has been located. Because of this 
geographic variation, it is inappropriate to use the same 
conversion factor for horseshoe crabs from all states. 
The conversion factor that was used by NOAA Fisher- 
ies (i.e., 1.21 kg per horseshoe crab) to estimate refer- 
ence period landing data does not accurately estimate 
total biomass. From our results, it seems that reference 
period landing data were overestimated, especially in 
cases where the fishery could have been male-biased. 
The effects of this inaccurate conversion factor could 
have been further magnified in areas where the aver- 
age size and weight of horseshoe crabs is much smaller 
than that for Mid-Atlantic states, notably embayments 
from the northern (Cape Cod to Maine) and southern 
(Gulf of Mexico) parts of the distribtuion range of this 
