266 PROCEEiDINGS OF THE NATIONAL MUSEUM vol.94 



Vandellia plazai Pellegrin, Bull. Mus. Hist. Nat. Paris, vol. 5, p. 158, 1899 

 (Apure River, Venezuela). 



Subfamily Tridentinae 



KEY TO THE GENERA OF THE SUBFAMILY TRIDENTINAE 



la. Opercular and interopercular patches of spines confluent; several series of 

 strong teeth in each jaw; gill membranes broadly united with isthmus, 



without a free margin Miuroglanis » Eigenmann and Eigenmann 



16. Opercular and interopercular patches of spines distinct and separately mov- 

 able; eyes lateral, seen as well from below as from above; gill membranes 

 joined across isthmus, forming a wide free fold. 

 2a. Opercle with 10 curved spines; interopercle with 3 or 4 smaller but similar 

 spines; depth 13, head 9, in standard length; 2 maxillary barbels present but 

 minute; nasal barbels absent; branchiostegals 4; dorsal rays 10 to 12; 

 anal 20 to 25; pelvic i, 2, minute; pectoral i, 4; caudal said to be 



rounded Tridens '^ Eigenmann and Eigenmann 



26. Opercular spines G to 10; interopercular spines 4 to 8; depth 4 to 8; head 5 

 to 6>^ in standard length; teeth curved and in 3 rows in upper jaw. 

 3a. Opercular spines 10; interopercular spines 8; 2 maxillary barbels well 

 developed; nasal barbel present or absent; dorsal rays 7 to 10; anal 17 



to 21; pelvic probably i, 4 Tridentopsis i« Myers 



36. Opercular spines 6; interopercular spines 4 to 6, usually 6; dorsal rays 

 9 to 11; anal 20 to 23; nasal barbel absent; brancliiostegals 5; pelvics 

 i, 4; pectorals i, 5 Tridensimilis, new genus 



TRIDENSIMILIS, new genus 



Eigenmann and Eigemnann (Proc. California Acad. Sci., ser. 2, vol. 

 2, pp. 53-54, 1889) described the genus Tridens, based on T. melanops 

 (genotype) and represented by 27 specimens from Iga, Amazons. 

 One of the paratypes, U.S.N.M. No. 41522, was sent to the National 

 Museum and has been studied by me. At the same tune, Eigenmann 

 and Eigenmann {op. cit., p. 54) described Tridens breiis from a smgle 

 specimen taken at Tabatinga, but this type was lost years ago at the 

 Museum of Comparative Zoology. Myers (Copeia, No. 148, p. 84, 

 1925) described the genus Tridentopsis, based on T. pearsoni (genotype) 

 and recorded previously by Pearson (Indiana Univ. Studies, No. 64, 

 pp. 17-18, 1924) as Tridens brei-is from Lake Rogoagua, Bolivia. 

 Miss La Monte (Amer. Mus. Nov., No. 1024, p. 1, 1939) described 

 Tridentopsis tocantinsi from the Rio Tocantins, northeastern Brazil. 

 Obviously her species is close to pearsoni, except in the absence of 

 nasal barbels. 



It should be noted that up to the present time all species referred to 

 the Tridentinae have come from the Amazon Basin. However, I had 



H Genotype, Miuroglanis platycephalus Eigenmann and Eigenmann, Proc. California Acad. Sci., vol. 2, 

 r. 66, 1889 (Jutahy). 



n Genotype, Tridens melanops Eigenmann and Eigenmann, Proc. California Acad. Sci., vol. 2, p. 83, 

 18S9 (lea). Most of these data are from U.S.N.M. No. 41522, one of the paratypes of T. melanops. 



i« Genotype, Tridentopsis pearsoni Myers, Copeia, No. 148, p. 84, l<J2o, based on Tridens brevis Pearson 

 (not of Eigenmann and Eigenmann) (Lake Rogoagua, 'BoU\ia).— Trid(ntopsisiocontinsi La Monte, Amer. 

 Mus. Nov., No. 1024, p. 1, 1939 (Rio Tocantins, Brazil). 



