280 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



PROPORTIONAL MEASUREMENTS 



In conventional taxonomic accounts of the 

 species of paralichthyx and related species a 

 prominent part of the description is usually com- 

 prised of statements of the proportional meas- 

 urements of various parts, while the statements 

 referring to gill raker, fin ray and scale counts are 

 apparently based on a few specimens; and such 

 important characters as the sti-uctiu'e of the scales, 

 ctenoid or cycloid, and the presence or absence of 

 accessory scales are often left out. However, for 

 the purpose of distinguishing the species properly, 

 proportional measurements are of secondary im- 

 portance, except in a very few cases, the essential 

 requisites being frequency distribution tables of 

 the numbers of gill rakers, fin rays and scales, and 

 descriijtions of the typical structure of the scales 

 and the presence or absence of accessory scales. 

 In this investigation four measurements, the great- 

 est depth, the length of the maxillary, the length 

 of the head, and the interorbital width, were 

 studied in detail, in order to test the practical value 

 of these characters in the proper distinction of tlie 

 species. The length of the pectoral diifers as be- 

 tween woohnani and brasiliensis, and it may pos- 

 sibly show average diti'eiences between some other 

 species. However, it was not investigated in de- 

 tail, since after a cursory examination it was de- 

 cided that it would not be of a more decisive na- 

 ture than the other four measurements. In the 

 genus Hippof/losxinn the upjier eyeball and orbit 

 were also measured, because these measurements 

 are of some importance in separating the sub- 

 genera; while in the subgenus Hii>po(/lo><shia the 

 greatest width of the maxillary is of some im- 

 portance in distinguishing the species. 



There may be other measurements showing 

 average differences between the species, but if there 

 are any. they are ap[)arently not pronounced. It 

 was noted, as was to be expected, that the pro- 

 portions of the parts investigated differ greatly 

 with size, and the data were consequently sep- 

 arated by size groups. Indeed, the intraspecific 

 differences due to size are frequently greater than 

 the interspecific differences. The results of these 

 measurements, segregated by size groups, are con- 

 veniently presented in tables 7 and 8. Sumuiariz- 

 ing the results, it may be stated that they are of 

 some value, as follows. 



The greatest depth is useful in separating ad- 

 sprrsuH from califomicus and aestVMrius. This 

 difference is of no practical value in the identifica- 

 tion of specimens because the former species is 

 geographically discontinuous with the latter two 

 and occurs in a widely separated region. How- 

 ever, since the other specific characters separating 

 adspeisvs from the other two species are now 

 shown to intergrade to a very large extent, the 

 difference in the greatest depth is useful in prov- 

 ing their distinctness. The relative depth is also 

 useful in aiding the sejiaration of .^(/i/armle?ifiix 

 from lefhoKfigina and d( iifatux. In this case it is 

 especially important to compare specimens of ap- 

 proximately the same size, and the depth is not the 

 only important distinguishing character. Other 

 characters show nearly as much divergence as the 

 relative de]ith, although there is more or less in- 

 tergradation in every one of those characters. 



The relative length of the head and that of 

 the maxillary are useful in separating woohnani 

 fiom hra sinensis. While there is some intergra- 

 dation in this character between the two species, 

 the degree of ovei'lapping is apparently less than 

 in the other characters separating them. Here 

 again it is necessary to compare individuals of ap- 

 proximately like size, and this character is of no 

 jn-actical value in the identification of specimens, 

 the two species being discontinuous geographi- 

 cally, the former occurring on the Pacific coast and 

 the latter on the Atlantic coast. 



The interorbital width is useful in separating 

 lethostigma. from dentatus in specimens over 10(1 

 nun. long. In this .case, also, it is necessary to 

 compare individuals of approximately like size: 

 while, on the other hand, the difference in the num- 

 ber of gill rakers distinguishes these two species 

 readily. 



Briefly then, of the proportional measurements 

 investigated in detail, namely, the depth, head, 

 maxillai-y, and interorbital, the depth of body is 

 of value in aiding the separation of squamilentus 

 and adspers-us from related species; while the 

 length of the head and ma.xillary is useful in pre- 

 senting evidence that woohnam and brasiliensis 

 are distinct species. H. mi/stacivm may be dis- 

 tinguished from H. stomata by differences in the 

 length and width of maxillary and the head length. 

 In other species, differences in these measurements 

 are of lesser im]iortnnce. 



