STAEFISHES OF THE PHILIPPINE SEAS. 379 



able to compare several species, including the types of Perrier's N. 

 gomophia and N. obtusa. The latter is small and, I think, immature. 

 It seems to be scarcely comparable with large specimens of other 

 species, Perrier compares it with variolafa, but I could not see a 

 close resemblance. Its relationship is with the strongly tuberculate 

 forms, such as frianti and twinulosa. Perrier has also compared his 

 gomophia with aegyptiaca^ a comparison which threw me entirely 

 " off the track " until I saw the type. It does not resemble aegyptiaca 

 at all but closely resembles novae-caledoniae, judging by the speci- 

 mens of the latter in the British Museum, which Perrier himself men- 

 tions (1875, p. 164) and presumably identified. 



Dr. H. L. Clark kindly loaned me a number of specimens which 

 he collected at Mer (or Murray) Island, Torres Strait. These have 

 been most valuable, and three are figured on plate 108 together with 

 a specimen of variolata (No. 838, Museum of Comparative Zoology) 

 from Mauritius. I am not at all certain of the identification of the 

 specimen figured on plate 108, figure 2, which I have called novae- 

 caledoniae. Without a photograph of the type, or an authentic speci- 

 men for comparison, any identification of this species must be re- 

 garded as tentative. 



The few notes I was able to make in London are included in the 

 following synopsis, which is in no sense intended as a revision of 

 the forms studied, since much time and material will be required 

 for that. I have suggested that N. -finschi is synonymous . with N. 

 pauciforis. N. ohtusa is not included in the key, since it is too young 

 to be comparable with adult examples of even its nearest relatives. 

 Only such species as I have seen are in the synopsis. 



SYNOPSIS OF SOME INDO-PACIFIC SPECIES OF NAEDOA. 



o \ Abactinal plates very slightly convex to markedly convex, but not high 

 enough to be considered hemispherical or tuberculate. 

 h \ Abactinal plates forming very regular longitudinal and transverse series ; 



furrow spines 6 or 7__ semiregularis, p. 383. 



I) '. Abactinal plates not regularly arranged in longitudinal and transverse 

 series. 

 c\ Abactinal plates very much larger than the papular areas, elliptical in 

 contour and usually with long axis of the ellipse transversely ori- 

 ented. In specimen with R 68 mm. there are 5 longitudinal series of 

 large alternating plates between the 2 rows of superomargiuals ; 

 adambuUicral spinelets in 3 series, viz., 4 (or 3), 3, 3, the last granuli- 

 form. Granules surmounting the depressed convex plates nearly 

 uniform in size, polygonal, close-set, and 2 or 3 times the size of the 

 the granules in the depressions between the plates.* 



variolatus Gray (pi. 108, fig. 4). 



1 From Gray's specimen, Mauritius. The rays are shorter and blunter than In any 

 other species in this section, and are not so tapered as those of species under C*. 

 R=68 mm., r=13 or 14 mm., R==about 6 r. The specimen figured (pi. 108, fig. 4) is 

 from Mauritius (No. 838, Mus. Comp. ZoiJl.) 



