PEREZ FARFANTE: DIAGNOSTIC CHARACTERS IN HEPOMADUS 



Figure 8. — Hepomadus lener. Thelycum, ^ 47 mm c.l. 

 offBackBay, Va. 



The other record is that of a female reported 

 by Ramadan (1938) from off Zanzibar [Tan- 

 zania], in 2,930 m; this record leaves no doubt 

 as to the presence of H. tener in the western 

 Indian Ocean. 



DISCUSSION 



Previous authors (Smith, 1886; Bouvier, 

 1908; Burkenroad, 1936; Roberts and Pequeg- 

 nat, 1970) who have reported H. tener from the 

 western Atlantic have cited a number of dif- 

 ferences between their material and specimens 

 that had been described earlier. In addition to 

 these western Atlantic records of this species, 

 Milne Edwards and Bouvier (1909) identified a 

 male from SW of Dry Tortugas Islands as 

 H. glacialis on the basis of several features, 

 including the presence of a well-defined spine 

 on the midposterior margin of the fourth and 

 fifth pleonic terga but, at the same time, pointed 

 out other characters that distinguish their 

 specimen from the type of H. glacialis. 



Burkenroad (1936) presented a detailed dis- 



cussion of the differences in various characters 

 (e.g., shape of the proximal projection, "inner 

 lobe," of the second segment of the mandibular 

 palp, relative lengths of the stylocerite and the 

 scaphocerite, development of the postrostral 

 carina, length of the spine at the midposterior 

 margin of the third pleonic tergum) among 

 known specimens of H. tener, and interpreted 

 the differences as individual variations or those 

 due to differential growth rates. He also noted 

 that the small male assigned to H. glacialis 

 by Milne Edwards and Bouvier differs from 

 H. tener (according to their description), in 

 possessing a larger rostrum, a smaller interval 

 between the last two rostral teeth, a shorter 

 spine on the third pleonic tergum, an obtuse 

 posteroventral tooth on each side of the sixth 

 pleonic somite, and in lacking setae on the 

 upper side of the endite of the basis of the 

 second maxilla. He concluded that if such dif- 

 ferences exist, the small male belongs to a 

 species other than H. toier. Furthermore, 

 Burkenroad stated that his specimens had no 

 trace of a spine at the midposterior margin of 

 the fourth and fifth pleonic terga, and that the 

 presence of those spines in the male reported 

 by Milne Edwards and Bouvier is the strongest 

 reason for identifying the latter as H. glacialis. 

 Later, Ramadan (1938) after examining the 

 type of H. glacialis, cited three features by 

 which Milne Edwards' and Bouvier's specimen 

 differs from the type, and suggested that it 

 probably represents an undescribed species. 



The study of collections at my disposal as 

 well as data presented in previous descriptions 

 of other material indicate that all known 

 western Atlantic Hepomadus are different 

 from H. glacialis. During the course of the 

 investigation it also has become evident that a 

 considerable degree of variation exists among 

 specimens from the region. In each of the 

 characters considered, however, there is a broad 

 range of variation, which seems to indicate that 

 all the specimens belong to a single species, 

 Hepomadus toier. 



The principal differences between H. glacialis, 

 the holotype of which has been made available 

 to me, and H. tener are given in Table 1. 



To analyze the various characters considered 

 by previous authors to be of possible diagnostic 



449 



