LEIS: SYSTEMATICS AND ZOOGEOGRAPHY OF THE PORCUPINEFISHES 



Diodon holocanthus eggs and larvae have 

 been found in Hawaiian waters from February 

 through September, with an apparent peak in 

 abundance in May-June, although they are 

 never common. Larvae usually occurred singly 

 in plankton tows ( volume filtered 200- 1 ,000 m-'^). 

 Although as many as 30 eggs 1.000 m'^ have 

 been taken, 1-5 eggs/1,000 m^ were more usual, 

 and most tows contained none. Eggs were usu- 

 ally found close to shore, but larvae rarely were 

 found closer than 1 km from shore (pers. ob- 

 serv.). 



Holotype. — No holotype or type-series is known to 

 exist. Linnaeus based his description on that of 

 Artedi (1738). 



Distribution . — Diodon holocanthus is circumtrop- 

 ical in distribution, but is seemingly absent in the 

 southwest and central Pacific east of the andesite 

 line (the separation of continental from oceanic 

 rocks, Figure 14). However, it reappears in 

 Hawaii, Pitcairn, and Easter Islands. Cuvier's 

 holotype of D. quadrimaculatus was allegedly col- 

 lected by Peron in Tahiti (see Le Danois 1961). 

 Inasmuch as it is known that much of the locality 

 data accompanying Peron's specimens are incor- 

 rect (associated with a shipwreck, see Whitley 

 1931:25) this record is questionable. There is evi- 

 dence of divergence of the Atlantic population! s) 

 from those of the Indo-Pacific (see Remarks). 



Remarks. — I follow the spelling holocanthus 

 (rather than holocanthus of many authors) which 

 was used consistently by both Linnaeus and Ar- 

 tedi (see also Bailey et al. 1970), and is thus not 

 considered to be a misprint as maintained by Jor- 

 dan and Evermann (1891). Linnaeus' description 

 is brief; the only useful information being the 

 statement that the spines are terete and ex- 

 tremely long on the head and nape. However, this 

 can apply only to D. nicthemerus or D. holocan- 

 thus. Assuming that "Habitat in India" means 

 India as understood today, and not the entire 

 Indo-Pacific, D. nicthemerus is eliminated. How- 

 ever, even if "Habitat in India" means the entire 

 Indo-Pacific, it is unlikely that specimens of D. 

 nicthemerus, a species apparently confined to 

 southern Australia, could have reached Artedi by 

 1738. In any case, subsequent usage and stability 

 demand that the name D. holocanthus apply to the 

 species described above. 



Diodon pilosus is synonymized with D. holocan- 

 thus on the basis of Mitchill's observation that no 

 spines were present between the dorsal and caudal 

 fins of his small (ca. 38 mm) New York specimen. 

 Diodon holocanthus is the only Atlantic species 

 that lacks peduncle spines. Mitchill distinguished 

 D. pilosus on the basis of its flxible spines, but this 

 is the usual condition in small specimens. No 

 holotype is known to exist. 



Cuvier's types are extant. Information and 

 photographs of these specimens (catalog numbers 

 and other information are given by Le Danois 

 1961) provided by M. L. Bauchot (pers. commun., 

 MNHN, 20 May 1975) clearly establish D. 

 novemmaculatus, sex?naculatus . quadrimac- 

 ulatus, and multimaculatus (all of Cuvier) as 

 junior synonyms of D. holocanthus . Inasmuch as 

 Cuvier's (1818) descriptions are relatively clear, 

 only his D. novemmaculatus requires comment. 

 The holotype of D. novemmaculatus (MNHN 

 A. 9928, 107 mm) is D. holocanthus, apparently 

 from the Atlantic (no locality data are available 

 for this specimen). A spine is present below the 

 anterior margin of the eye and the eye bar is dis- 

 continuous over the interorbital. Unfortunately, 

 Cuvier's figure resembles D. liturosus as much as 

 D. holocanthus (the figure shows the frontal spines 

 shorter than they actually are). This probably led 

 Bleeker (1865) to apply the name D. novem- 

 maculatus to D. liturosus. 



Diodon maculifer Kaup ( 1855) is included here 

 with some questions. Kaup's description is of little 

 help, and no type material can be found in the 

 British Museum where it would be expected to 

 reside. The holotype may have been part of Kaup's 

 lost personal collection (A. C. Wheeler, pers. com- 

 mun.). Examination of one of the South African 

 (Kaup's type-locality) specimens of "Diodon 

 maculifer" listed by Giinther (1870) (BMNH 

 1845.7.3.103, 100 mm, loaned by A. C. Wheeler) 

 reveals it to be an inflated, dried D. holocanthus. 

 In this specimen, inflation is so great ( an artifact of 

 stuffing and drying?) that the subdermal spine 

 bases project through the dried skin. Thus, the 

 base of the spines appear to be expanded and 

 transversely compressed. The only characteristic 

 feature of Kaup's description is the compressed 

 nature of the spines, and it seems likely that his 

 description was based on a dried, inflated D. 

 holocanthus. 



Steindachner's Atopomycterus bocagei can be 

 placed in the synonomy of D. holocanthus on the 

 basis of information on the holotype (NMV 63848) 



563 



