228 



Fishery Bulletin 90(2), 1992 



These data indicate that the annuli become discern- 

 ible in May. Because these rings are present on juve- 

 niles as well as adults, they appear to be related to 

 seasonal patterns of growth rather than reproduction. 

 The annuli are difficult to see when they are at the very 

 edge of the vertebral centra. For this reason, they are 

 probably not detected until some additional growth has 

 occurred after they are laid down. Yasuda (1940) has 

 shown that on vertebrae ofScombrops sp. annuli were 

 formed 1.5 months later than on the otoliths. So it is 

 likely that the annuli (composed of a step and a translu- 

 cent band) found on goosefish vertebrae represent the 

 end of fast growth (the step) in late-fall and a period 

 of slow winter growth (the translucent band). 



While several authors have studied growth in L. pis- 

 catoritis and L. budegassa (Fulton 1903, Guillou and 

 Njock 1978, Tsimenidis and Ondrias 1980), only Con- 

 nolly (1920) has looked at growth in L. americaniis. He 

 based his growth estimates on vertebral annuli counts, 

 but his sample size was only six individuals. His results 

 were as follows: age 1, 114 mm; age 4, 457 mm; age 

 8, 737mm; age 9, 787mm; age 10, 940mm; age 12, 

 1016mm. These estimates are slightly lower than found 

 in this study, but a slower growth rate would be ex- 

 pected in the colder Canadian waters in which Connolly 

 conducted his study. 



The growth rate of L. americanus is intermediate 

 to L. piscatorius and L. budegassa. Figure 12 compares 

 the mean back-calculated lengths for the two European 

 species (from Tsimenidis and Ondrias 1980) with data 

 presented here for L. americanus. 



The differences in observed and back-calculated 

 mean lengths between males and females past age 4 

 are small, but appear to be real. This is the most com- 

 mon form of sexual dimorphism among fishes (Moyle 

 and Cech 1982). Tsimenidis and Ondrias (1980) found 

 similar small differences between the sexes for L. bu- 

 degassa and L. piscatorius. 



More significant is the difference in mortality be- 

 tween the sexes implied by the data. The heavier mor- 

 tality of males may be caused by increased predation 

 due to their smaller size, but this does not seem likely. 

 Perhaps the males exhibit behavioral or distributional 

 differences which make them more susceptible to 

 predation or fishing effort. A final possibility is that 

 they simply reach senescence before females. 



The von Bertalanffy growth equations fit the back- 

 calculated lengths extremely well. The values for L^ 

 for both sexes seem somewhat inflated. The maximum 

 reported size for L. americanus is approximately 

 1220 mm (Bigelow and Schroeder 1953). The largest 

 female collected in this study was 1115 mm and the 

 calculated L^ was 1576 mm. The largest male collected 

 was 900 mm compared with a calculated L of 1018- 

 1460 mm. The inflation of L„ is caused by a lack of 



representatives from the older age-classes. This is a 

 common problem in age and growth studies. The 

 asymptotic length is therefore not well defined for 

 either sex in this study. The sampling effort was be- 

 lieved to be intense enough to sample these larger 

 individuals if they were present in the population. It 

 is concluded that these individuals are simply not pres- 

 ent. This may be the result of commercial fishing 

 pressure (groundfishing and scalloping), which tends 

 to be selective towards larger individuals. 



Acknowledgments 



We are indebted to all the graduate students and staff 

 members at the Virginia Institute of Marine Science 

 who assisted at various points in this study. D. Sved, 

 M. Chittenden, and W. DuPaul provided helpful re- 

 views of this manuscript. Ship time was provided by 

 the Northeast Fisheries Science Center and the fishing 

 vessels Captain Wool, Vi7'ginia Queen, Virginia 

 Cavalier, and Cara Lyn. Funding was provided by Sea 

 Grant, National Oceanic and Atmospheric Administra- 

 tion, U.S. Department of Commerce, under Grant 

 NA86AA-D-SG042, through the Virginia Sea Grant 

 Program, Project RC/F-10, J.A. Musick, Principal In- 

 vestigator. This manuscript was based on a thesis 

 submitted by the senior author in partial fulfillment of 

 the M.A. degree, School of Marine Science, College of 

 William and Mary. 



