NOTE Kohler et al.: Length-weight relationships for I 3 species of sharks 



413 



included in the analysis because of the shark's un- 

 usual size. Data were obtained opportunistically 

 throughout each year but most (88%) were collected 

 in the months of June, July, and August off the north- 

 east United States between North Carolina and 

 Massachusetts. Only lengths and weights measured 

 by the authors and other members of the API or by 

 cooperating biologists are included in this report. 

 Measurements of embryos and fish known to be preg- 

 nant were excluded from the data set. 



All lengths were measured with a metal measur- 

 ing tape to the nearest centimeter in a straight line 

 along the body axis; the caudal fin was placed in a 

 natural position. Fork length (FL) was measured 

 from the tip of the snout to the fork of the tail. Total 

 length (TL) is defined as the distance from the snout 

 to a point on the horizontal axis intersecting a per- 

 pendicular line extending downward from the tip of 

 the upper caudal lobe to form a right angle. 



Total weight (WT) of each shark was measured to 

 the nearest pound and converted to kilograms. The 

 majority of the fish were weighed hanging from the 

 caudal peduncle which allowed any water in the 

 stomach, and in some cases stomach contents, to drop 

 out prior to weighing. Many fish were examined in- 

 ternally and, if unusually large amounts of water or 

 contents were found in the stomach or abdominal 

 cavity, the weights were subtracted to obtain a more 

 accurate weight. 



Fork-total length relationships for 13 species of 

 shark (n =5,065) were determined by the method of 

 least squares to fit a simple linear regression model. 

 Linear regressions of fork-to-total length were cal- 

 culated with their corresponding regression coeffi- 

 cients, sample sizes, and mean lengths. These data 

 were then combined into four family groups: 

 Alopiidae (thresher sharks), Lamnidae (mackerel 

 sharks), Carcharhinidae (requiem sharks), and 

 Sphyrnidae (hammerhead sharks), and linear regres- 

 sions and TL/FL percentages were calculated for each 

 group. 



An allometric length-weight equation was calcu- 

 lated by using the method of Pienaar and Thomson 

 (1969) for fitting a nonlinear regression model by 

 least squares. The form of the equation is WT=(a)FL h , 

 where WT=total weight (kg), FL=fork length (cm), 

 and a and b are constants. Length-weight relation- 

 ships, mean lengths and weights, and size ranges 

 were determined for 13 species of sharks (rc=9,512). 

 Literature values for maximum fork length and fork 

 length at maturity were also included. The regres- 

 sions of the length- weight equations expressed loga- 

 rithmically were tested for possible significant dif- 

 ferences (P<0.05) between males and females by using 

 an analysis of covariance test for homogeneity of slopes. 



Fork length is used throughout this report as the 

 basis for all conversions and comparisons. We have 

 found fork length to be a more precise measurement. 

 For comparative purposes, all values published else- 

 where as total lengths were converted to fork lengths 

 by using the species' equations presented in this 

 paper. 



Minimum sizes at maturity reported here are from 

 published accounts with their original sources refer- 

 enced, with the exception of Alopias vulpinus and 

 Carcharodon carcharias. Minimum size at maturity 

 for the thresher shark and the male white shark were 

 determined by Pratt 1 who used the following crite- 

 ria: smallest male with calcified claspers that rotate 

 at the base and smallest gravid female. When con- 

 siderable variation occurred among published ac- 

 counts, traditional sizes at maturity were chosen 

 primarily from Atlantic populations. The maximum 

 sizes and maximum sizes at birth used here are sum- 

 marized in Pratt and Casey (1990). 



Results and discussion 



Linear regressions of fork-to-total length were cal- 

 culated for 13 species of shark and four family groups 

 (Table 1). The slopes of the regression lines of the 

 four families decrease with the increasing length of 

 the upper caudal lobe. The mackerel sharks have 

 lunate tails with the upper and lower caudal lobes 

 almost equal in size. The requiem sharks, hammer- 

 head, and thresher sharks have heterocercal tails 

 with the upper lobe longer than the lower. The latter 

 group have very long upper caudal lobes with the 

 fork length approximately 60% of the total length. 

 The fork length represents 92%, 84%, and 77% of 

 the total length for the mackerel, requiem, and ham- 

 merhead sharks, respectively. 



A total of 9,512 sharks representing 13 species 

 were measured, sexed, and weighed. There were no 

 significant differences in slope or intercept of the 

 length-weight relationships between males and fe- 

 males for any of the species and therefore one equa- 

 tion, calculated with the sexes combined, was used 

 to represent the data for each species (Table 2). 



Size at maturity for males and females is difficult 

 to determine for pelagic sharks and can vary in dif- 

 ferent parts of the world (Pratt and Casey, 1990). 

 The discrepancy is due, in part, to the use of vari- 

 able criteria in determining a precise length at sexual 

 maturity (Springer, 1960; Clark and von Schmidt, 

 1965; Pratt, 1979) and thus maturity is often reported 



1 H. L. Pratt, National Marine Fisheries Service, Narragansett, 

 RI 02882. Pers. commun., May 1993. 



