378 Travsactions. 



Atypichthys strigatus, Gimther. 



Of this widely distributed species we have three examples from the 

 Kermadec Islands. Mr. Oliver states that they are common among rocks 

 all round the coast. 



• 



Goniistius gibbosus, Richardson. 



The distinguishing features of the members of the genus Goniistius have 

 not yet been clearly expressed, and aAvail investigation at the hands of some 

 worker with a representative series. The only example received from the 

 islands is in fair agreement with Steindachner's figure^* but the oblique 

 bands are not so well defined, and the last, or eighth, band does not run 

 on to the tail ; the anal fin, also, is black. The rays of the dorsal fin 

 number XVII, 33 ; those of the anal III, 8. The specimen measures 

 221 mm. in length. 



Chironemus marmoratus, Giinther. 



Several specimens, taken among rocks. 



Chromis, sp. 



The only member of the PoinacentridcF secured is a scaleless and other- 

 wise mutilated specimen found on the beach. It is not possible to identify 

 it specifically, but the following characters may be mentioned : Body less 

 than half the length ; second anal spine very long and strong, longer than 

 the longest dorsal spine ; pectoral longer than the head, and one-third the 

 length of the body ; tail long and deeply forked, its lobes acute ; the 

 caudal peduncle is slender, its depth one-third greater than the diameter 

 of the eye. It is not possible to ascertain the coloration, but traces of 

 orange exist about the opercles and at the bases of the fins. The length of 

 the specimen is 160 mm. 



Pseudolabrus inscriptus, Eichardson. 

 Previously recorded from Raoul Island. 



Pseudolabrus luculentus, Richardson. 



Represented by small examples taken from rock-pools. 



Cantherines analis, Waite. 



Pseudomonacanthus analis, Waite. Rec. Aust. Mus., v, 1904, p. 173, fig. 32. 



I associated with this species an example obtained at Sunday Island, 

 but noted differences in respect to the situation of the gill-opening and the 

 pectoral. The type being preserved in the Australian Museum, I forwarded 

 the specimen to Mr. McCulloch. who kindly writes as follows : " The speci- 

 men is undoubtedly Pseudomonacanthus analis. The type, as you will re- 

 member, is imperfect and beach-worn, and the positions of the gill-opening 

 and pectoral are consequently not quite correctly shown in the figure. The 

 small black spots seeii in the Kermadec Island specimen may be traced in 

 the type." 



In my paper I made comparison with P. degeni, Regan.t and now, 

 having an example for direct comparison, Mr. McCulloch writes, " P. degeni 

 is much more slender than P. analis, the caudal peduncle is longer, and the 

 spine is over the hinder margin, and not over the middle of the eye." 



* .Steindachner and Doderlein, Fische Japans, ii, 1883, pi. vii, fig. 2. 

 t Regan, Pi-oc. Zool. Soc, 1902. ]!. 299, }>]. xxiv, fig. 1. 



