64 



HIALMAR RENDAHL 



of eye; eye 3,9 in head; interorbital space i 9 eye diameters and 0,6 length 

 of snout; longest (5***) dorsal spine 2,5 in head and 1,1 in longest dorsal ray; 

 second anal spine 2,4 in head. I could not detect any ciliated scales but only 

 cycloid ones. 



8. Girellops nebulosus Kendall and Radcliffe. — Girella nebulosa 

 Kendall and Radcliffe 1. c. p. 120, pi. Ill, f. 2 and pi. IV, f. i and 2; Girellops 

 nebulosus Regan 1. c. p. 369, pi. LVII; FUENTES 1. c. p. 16, lam. V. 



Three specimens, caught in a pool on the shore; ']6^ 53, and 38 mm (s. c.) 

 The present samples agree entirely with the descriptions and figures 

 published by the previous authors. As stated by Kendall and RADCLIFFE the 

 teeth are moveable. 



9. Thalassoma umbrostygiiia Riipp. — Kendall and Radcliffe, 

 1. c. p. 141. — Julis uinbrostigma Regan, 1. c. p. 371 and FuENTES, 1. c. p. 

 22, lam. IX. 



Native name: Paohu. — Three specimens, 248,5, 219 and 193 mm (s. c). 



These specimens agree with Rufpell's figure (Neue Wirbelthiere zu der 

 Fauna von Abyssinien gehorig. Fische des rothen Meeres. 1835. PI. 3, Fig. 2), 

 except that all or nearly all scales of the back and sides of the body have 

 blackish basal spots and that the opercles and cheeks are more richly spotted, 

 as the photograph in FuENTES' paper shows. 



To Jordan's statement in the diagnosis for this species (Bull. U. S. Fish 

 Comm. vol. XXIII, pt. I, 1903, p. 300), that the eye is contained 6,6 times in 

 head, a character also used by him in his table for examination of the 

 Hawaiian species (1. c. p. 295), I shall remark that in my specimens this pro- 

 portion varies between 8,6 and 7,0. Evidently it is impossible to ascribe to 

 the relative dimension in question such a high value for distinguishing Th. 

 purpureum and utnbrostygina from each other as Jordan has done. 



In the table below the dimensions of my specimens are recorded: 



As to the nomenclature I have followed the spelling umbrostygma as 

 used by Ruppell in his original description (1. c. p. ii); later, the name has 

 been wrongly quoted as timbrostigina by GuNTHER and the following writers 

 in their references to ROppel. 



10. Labrichthys fuentesi Regan 1. c. p. 371, pi. V. f. i, FuENTES 

 1. c. p. 22, lam. VII, f. 2. — Pseudolabrus inscriptus Kendall and Radcliffe 1. c. 

 p. 137, pi. V, f. 2 and pi. VI, f. i. 



