6o HIALMAR RENDAHL 



island of the Paumotu archipelago, naturally makes it a matter of great interest 

 to enter upon the relations of the fish fauna. Kendall and Radcliffe 

 remark that »the fauna of Easter Island seems to be different from that of its 

 nearest neighbor, Manga Reva (Paumotus) and appears closest to that of Nor- 

 folk Island*. Regan points out that it is not related to the Chilean fauna 

 and also calls attention to the resemblance to the fauna of Norfolk Island. 



My own studies have led to the result that I entirely agree with Regan 

 that there is no element related to the Chilean or, as far as is known, the 

 South American Pacific fauna. On the other hand, from a zoogeographical 

 point of view I can find no reason to proclaim a closer and restricted rela- 

 tionship with the coastal fauna of the distant Norfolk Island. 



If we examine the Easter Island species with regard to their general 

 distribution, we can divide them into the following groups^: 



A. Species of wide distribution, outside the tropical Pacific at least 

 also found in some part of the Indian Ocean. 



Myripristis pralinius C. V. PlatopJirys inancus Brouss. 



Kyphosus cinerascens Forsk. Parupencus trifasciatus Lac. 



Cheilio inermis Forsk. Xanthichthys lineopunctattis Hollard. 



Thalassoma piirpiireiim Forsk. Belone platyura Bennet. 



» umbrostygjua Riipp. Diodon holacanthus Lin. 



Salarias arenatus Bleek. 



B. Polynesian species. 



HolocentruDi punctatissimum C. V. Paumotu, Samoa, Hawaii. 



Anampses pidcher Regan. Tahiti. 



Alticus varioloms C. V. Samoa, Guam, Tonga, Hawaii. 



Pomacentrus inornatus Regan. PSamoa. 



Decapterus sanct(B-helencB C. V. Distribution? 



C. Hawaiian species. 



Caranx cheilio Snyder. Teuthis umbra Jenkins. 



Kellogella oligolepis Jenkins. 



D. Norfolk Island species. 



Trachypoma inacr acanthus Giinth. 



E. Endemic species and their allies. 



Bathystethus orientale Regan, related to B. cultratum Forster, of Norfolk 

 Island. 



Kuhlia niitabiinda Kend. & Radcl, related to K. himilis De Vis, Queens- 

 land, Fiji, and to K. sandvicensis Steind., Hawaii, Society I. 



• I don't consider the two species determined by Kendall and Radcliffe as Mona- 

 canthus cirrifer Scbleg, and Gymnothorax dovii Giinth., the identification of which seems to be 

 very uncertain (cfr. Regan, p. 373). 



