YOUNG JACK CREVALLES 



419 



Although the perimeters were drawn liberally in an 

 attempt to estimate population variations from 

 the samples, some specimens may be expected to 

 exceed the estimates. Still, it provides a useful 

 treatment of these highly variable relationships. 

 In the composite graphs of these relationships 

 under Comparison of Species, page 426, the perime- 

 ter for each species is specifically symbolized. 



On the graphs illustrating the relation of third 

 dorsal spine length, first dorsal soft-ray length, 

 pectoral length, body depth at pelvic, head length, 

 eye diameter, and snout length to standard length, 

 each dot or other described mark represents the 

 coordinate of a single specimen. Arithmetical 

 plots of individual variates indicated that in most 

 instances the formula for a rectilinear regression, 

 Y = a-\-bX, adequately described the relation- 

 ships. The regression lines for the five common 

 species were determined by the method of least 

 squares over a designated size range of the speci- 

 mens represented in each graph. A line so deter- 

 mined is referred to in the text as the "calculated 

 regression line" and is illustrated in the graphs as 

 a solid line. The trend of such a line is frequently 

 extended on the graphs (as a line of dashes) to 

 allow an estimate of the relation of specimens 

 larger or smaller than those included in the calcu- 

 lated regression line. The composite graphs of 

 these relations under Comparison of Species con- 

 sist only of the calculated regression lines (specif- 

 ically symbolized for each species). Because of 

 the limitations of material and methods, only 

 certain size ranges are suitable for comparison. 



Use of the term "inflection," applied to the 

 descriptions of the relations of body parts follows 

 Martin (1949) and denotes a change in slope of the 

 relative growth line. 



FISH ILLUSTRATIONS 



Fish less than 16 mm. standard length were 

 drawn with a camera lucida. Fish above this 

 size were photographed, and an enlargement of 

 the negative was used to reproduce the outlines. 



NOMENCLATURE AND SYNONYMY 



The discussions of nomenclature point out, with 

 the exception of ruber and bartholomaei, that the 

 relationships of the Western Atlantic species of 

 Caranx to certain closely related forms from other 

 areas are uncertain; consequently, the nomen- 

 clature is tentative. Since authors have differed 



iu opinion concerning the nomenclature and 

 synonymy of forms ascribed to the genus Caranx, 

 a comprehensive review of the literature was neces- 

 sary to understand the problems involved. This 

 is briefly outlined under the species concerned. 

 The synonymies are mainly composed of only the 

 earliest reference to each applicable binomial that 

 I have found. The principal purpose is not to 

 revise the nomenclature, but to point out dis- 

 crepancies that exist and to stabilize the usage 

 until adequate revisions are made. 



THEODORE N. GILL COLLECTIONS 



All methods employed on the GUI cruises — in- 

 cluding plankton and dip-net collections, stomach 

 contents of species caught by trolling and hand 

 line, and network of collecting stations — are given 

 by Anderson, Gehringer, and Cohen (1956). The 

 following abbreviations apply to the Gill cruises 

 and specimens: Reg. (regular station), Spc. (spe- 

 cial station), Std. (standard station, east of Elbow 

 Cay, Bahamas), and TO (Tongue of the Ocean 

 station, Bahamas). 



DEFINITIONS 



The definitions of measurements and counts 

 and descriptions of terms are arranged in the same 

 sequence in which the characters are considered 

 in this and the following sections. Frequently, 

 descriptions of ontogenetic changes are included. 

 Since earlier workers have used varied, and often 

 inadequately described, methods, some explana- 

 tions in this section have been expanded for clari- 

 fication. 



BODY LENGTHS 



Standard length. — Measured from the tip of the 

 snout to the end of the hypural bones (the caudal 

 base). All body lengths referred to are in stand- 

 ard length unless otherwise noted. 



Fork length. — Measured from the tip of the snout 

 to the end of the shortest median caudal ray. 



Total length. — Measured from the tip of the 

 snout to a vertical at the tip of the longest lobe of 

 the caudal fin. 



Body length conversion factors. — The rectilinear 

 regression formula, Y=a-\-bX, was used to de- 

 termine the conversions of total length and fork 

 length to standard length. Standard length was 

 used as the independent variate so that fork 

 length =a + 6 (standard length); total length = a + 6 



