12 PROCEEDlls^GS OF THE XATION.VL MUSEUM vot. 84 



temporalis^ but I can not believe this to be the case, for several 

 reasons. My specimens all show the t3^pical coloration as figured 

 by Boulenger, the temporal band, the dark basal pectoral bar, the 

 mottled fins, and the plain brownish body color, although they all 

 have, in addition, a small, more or less distinct dark spot in the 

 middle of the caudal base. As for the difference in the lateral scale 

 count, I believe this to be due to the fact that Boulenger counted 

 some scales out beyond the hypural on the caudal base. In the six 

 specimens before me, the size of the mouth and the shape of the 

 head vary greatly. In the largest and the two smallest the maxil- 

 lary reaches about midway between the nostril and the eye; these 

 also have a less declivous snout profile. In the other three the 

 snout profile is more declivous and rounded and the maxillary 

 reaches variously almost to, just to, or slightly behind the anterior 

 part of the eye. The tricuspid lateral teeth form a difference of 

 some weight. However, the characters of T. temporalis have been 

 known only through eight or nine examples taken at relatively few 

 localities. I feel that when this and other species are known through 

 large samples from an adequate number of localities, it will be found 

 that dental and other variations will be somewhat greater than now 

 susi^ected. 



TELMATOCHROMIS BIFRENATUS. new specks 



Plate 1, A 



Holofype. — U.S.N.M. no. 84121, 40 mm in standard length, from 

 Kigoma, Lake Tanganyika; collected in February 1920 by H. C. 

 Raven. 



Paratype. — U.S.N.M. no. 102112, same size and locality as holo- 

 type. 



Only two species of this genus, first described by Boulenger, are 

 certainly known, although three forms of uncertain generic position 

 described by Steindachner (1909, pp. 400-404) in the genus Juli- 

 dochromis have to be considered. The present new form differs 

 widely, however, from each of Steindachner's unfigured species in 

 one or more important characters of squamation, fin count, body 

 proportions, or color. Further, it cannot be Jiilidochromis macro- 

 lepis Borodin (1931, p. 51; 1936, p. 21, pi. 1, fig. 5), which, from the 

 wretched figure, looks to me, as it did to Regan (1932, p. 28), like 

 a Lamprologus. 



Diagnosis. — Closely allied to T. vittatus Boulenger, but differing 

 distinctly in having a more elongate body, a longer head, a longer 

 and loss bluntly rounded snout, a larger and less inferior mouth, 

 a lower lateral scale count, a longer and more slender caudal pe- 

 duncle, and two lateral dark bands on the body instead of one. 



