Chase et al.: Reproductive life history of anadromous Osmerus mordax in Massachusetts 
159 
Figure 5 
Maturity ogive for rainbow smelt (Osmerus mordax) sampled in marine waters 
off Massachusetts in 2004-2007 (n=634), derived for sexes combined from a 
generalized linear model with binomial error and logit link functions. The 
dashed line indicates the length at which 50% of the population was mature, 
126 mm in total length. The gray area indicates the 95% confidence interval 
of the maturity ogive. The ogive coefficients are significant (P<0.001) for the 
intercept (standard error [SE] 4.226) and total length (SE 0.033). 
Table 4 
Summary of total lengths (TLs) of rainbow smelt (Osmerus mordax) for which age was determined 
in this study. Fish were sampled in Massachusetts in marine waters during a trawl survey in 
2004-2007 and in the Fore and Saugus Rivers with fyke nets in 2005-2007. Significant differences 
in TL among age and location groups were tested with the Kruskal-Wallis rank sum and multiple 
comparison tests (a=0.05). Asterisks (*) indicate groups with significantly different TLs within each 
age and sex category. 
Age 
Sex 
Sampling location or type 
in) 
Mean (mm) 
SE 
1 
Male 
Trawl survey 
0 
Fore River 
174 
140* 
0.88 
Saugus River 
149 
147* 
0.98 
2 
Male 
Trawl survey 
0 
Fore River 
382 
180* 
0.61 
Saugus River 
144 
189* 
1.12 
1 
Female 
Trawl survey 
19 
137* 
2.46 
Fore River 
77 
142 
1.45 
Saugus River 
51 
145* 
1.46 
2 
Female 
Trawl survey 
75 
184 
2.61 
Fore River 
159 
191 
1.16 
Saugus River 
39 
205* 
2.16 
All fish from the Saugus River were larger at age 1 and 
age 2 than fish from the Fore River and the trawl survey, 
except for age-1 female rainbow smelt from the Fore River 
(Table 4). Total lengths of age-1 and age-2 females from the 
Fore River were not different from those of females from 
the trawl survey. Year-to-year TL comparisons were lim¬ 
ited to the Fore River because of sample sizes. Differences 
were found among 3 comparisons of annual data for sam¬ 
ples from the Fore River: males caught in 2005 and 2007 
and females sampled in 2007 with females captured in 
both 2005 and 2006. These comparisons of length statis¬ 
tics may reflect changes in annual growth rates and cohort 
strength; however, these results can be confounded by 
pooling samples by year. 
