278 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



In alhiguttn, the three most prominent spots .ire 

 also ocellated and form an imaginary triangle, the 

 apex of which is also the prepeduncular spot, as 

 in cJriitatus ; the base, liowever, is situated cephalad 

 of the apex and is formed by the two anterior spots 

 of the upper and lower intermediate rows, respec- 

 tively. This will be designated as the large tri- 

 angle. Both of these imaginary triangles are pres- 

 ent to a greater or lesser extent in both species. 



In alhi(fiitta, the spots forming the large tri- 

 angle are the most jirominent and nearly always 

 present; those forming the small triangle are less 

 prominent, or faint, or absent altogether. In 

 dentahis, the spots in the small triangle are usually 

 the most pi-ominent and nearly always jiresent; 

 those in the large triangle are usually well marked, 

 but not so prominent as the others, often about as 

 prominent, sometimes rather faint. In letho- 

 stigma all spots are usually rather faint; some- 

 times the spots in the large triangle are somewhat 

 more prominent, but they are not ocellated. The 

 difference in coloration in the three common 

 species is thus not absolute, it consists of an un- 

 equal development in intensity of pigmentation of 

 different parts of the same color pattern. This 

 being the case, and considering also the variability 

 of intensity of pigmentation with individual fish, 

 it may readily be expected that specimens will 

 freciuently be encountered which could not be 

 placed by color alone. However, the majority of 

 specimens may be referred to their proper species 

 by color differences. As an illustration, the fol- 

 lowing test may be cited. A mixed lot of fish 

 consisting of dcnfnfus, alhigufta, and hthoMtigma, 

 from Beaufort, N. C, were separated by color. 



After the structural difl^erences were studied, it 

 was found that out of a total of 125 individuals 

 thus separated only 14 specimens had been re- 

 ferred to the wrong species; 11 deiitatuH were 

 placed in alhigidta. two of the latter were tenta- 

 tively identified with the former, and one Irfho- 

 sfigina was mistaken for an alhigutta. It should 

 be stated that this test was made before I had 

 much experience in discerning the color differ- 

 ences of the various species. It may thus be seen 

 that in practice, color is a valuable aid in the 

 proper identification of the species when used in 

 conjunction with the morphological differences, 

 although it is not altogether reliable by itself. 



The ty]iieal color of the other -species could not 

 be well determined with the available specimens, 

 but notes on the color of these specimens are given 

 under the separate accounts of the species. 



The spots in the five rows, in general, appear to 

 grow fainter with increased size. To a lesser ex- 

 tent this is also true of the most prominent spots, 

 and in very large specimens the typical specific 

 color pattern is often not discernible. 



A similar generalized color pattern is probably 

 present also in Pseudorhoi/ibvs and possibly also 

 in HippogJoxii'nia and other related genera, but 

 the material examined is insufficient to determine 

 this definitely for those two genera. In Il'ippo- 

 gloKnimi the most prominent spots are in two rows, 

 two or three spots in a row depending on the sub- 

 genus, and appear to be situated in the subdorsal 

 and supra-anal rows. In Pxeudorhomhus the loca- 

 tion of the most prominent spots differs with the 

 species. 



Taislk 7. — Piotjortiomil iiicusiircmoits of 3 species of Hippusliissiua 

 [Expressed as percentages of standard length] 



1 Measurements not including the soft scaleless border. 



