FISHERY BULLETIN; VOL. 76, NO. 3 



Although basically shorefishes, the diodontids 

 (at least Dwdon and Chilomycterus) are strongly 

 tied to the pelagic environment through pelagic 

 eggs and well-developed pelagic juvenile stages. 

 In Diodon these juveniles remain pelagic for 

 weeks or months ( judging from size) and are often 

 found far from shore. In fact, juvenile Diodon spp. 

 are commonly encountered in the stomachs of such 

 pelagic predators as dolphins (Gibbs and Collette 

 1959), and one species, D. eydouxii, is apparently 

 pelagic throughout its life cycle. 



The eggs of diodontids are poorly known. 

 Nichols and Breder (1926) described the unfer- 

 tilized eggs of Chilomycterus schoepfi from New 

 Jersey as demersal, nonadhesive, transparent, 

 and about 1.8 mm in diameter. However, Breder 

 and Clark (1947) suggested that the eggs of C. 

 schoepfi may be normally pelagic. The pelagic eggs 

 of D. holocanthus and D. hystrix from Hawaii were 

 briefly described as Diodon sp. and "diodontid II," 

 respectively, by Wat,son and Leis (1974). Sanzo 

 ( 1930) described the development of what are ap- 

 parently the pelagic eggs of D. hystrix (identified 

 as Crayracion sp.?) from the Red Sea. Wolfsheimer 

 (1957) reported an aquarium spawning of D. 

 holocanthus (identified by him as D. hystrix), but 

 provided little descriptive information on the 

 eggs. The eggs mentioned by Wolfscheimer sank, 

 but did not adhere, to the bottom of the aquarium. 

 They did not develop, .so it is likely that they were 

 not fertilized. 



Larval and juvenile Diodon are no better known 

 than the eggs. Blanco and Villadolid (1951) illus- 

 trated a juvenile "Diodon bleekeri" but this fish is 

 clearly a juvenile tetraodontid. Many juvenile tet- 

 raodontids have prominent spines, particularly on 

 the ventral surfaces. Fowler (1928) illustrated a 

 juvenile Diodon, identified as D. hystrix, but the 

 figure does not show spines on the caudal peduncle 

 (see below), so this identification is apparently 

 incorrect (assuming the drawing is accurate). No 

 locality or other descriptive data are given by 

 Fowler, so a specific identification cannot be made. 

 Sanzo ( 1930) illustrated two larvae that resulted 

 from rearing of his D. hystrix eggs and a juvenile 

 Diodon captured in a plankton tow. The illustra- 

 tion of this latter fish shows no peduncle spines, 

 but in other respects it resembles D. hystrix. Mito 

 ( 1966) illustrated a larval and a juvenile Diodon, 

 both identified as D. holocanthus. The pigmenta- 

 tion and the relatively small eye .shown in Mito's 

 illustrations more closely resemble the specimens 

 of D. hystrix studied here. At least four species of 



536 



Diodon occur in Japanese waters, and Mito's 

 specimens could be any of these, because he gives 

 no information as to how the identifications were 

 made. Nishimura (1960) reported on juvenile 

 Diodon cast ashore in the Sea of Japan, but did not 

 provide specific identifications. 



MATERIALS AND METHODS 



Measurements and counts are as defined by 

 Hubbs and Lagler (1958:19-28) unless otherwise 

 stated. Measurements routinely were made with 

 needle point dividers to the nearest 0.5 mm. Fish 

 <10 mm and all eggs were measured under a dis- 

 secting microscope to the nearest division of the 

 ocular micrometer ( ±0.02 mm at 50 x, the power 

 normally used). All measurements are from pre- 

 served specimens. 



Unspecified lengths are in millimeters standard 

 length. Caudal peduncle length was measured 

 from the posterior base of the dorsal fin to the end 

 of the hypural plate. Head width was measured 

 immediately behind the eyes. Body width was 

 measured at the base of the pectoral fin. Width of 

 the eye was taken horizontally across the clear 

 cornea. Measurements are given as proportions of 

 standard length. 



Dorsal and anal fin ray counts included all rays, 

 branched and unbranched. The last two rays were 

 counted separately because they have separate 

 bases. Pectoral fin ray counts excluded the upper 

 ray. This ray, although well developed in small 

 ( <30 mm) juveniles, is a rudiment in adults and is 

 often not visible because it is embedded. In large 

 specimens, the fin bases are especially fleshy and 

 accurate fin ray counts are difficult to make with- 

 out dissection or radiography. 



Body measurements are given as range, mean 

 (I), and standard deviation (SD). The sample size 

 for the measurements is given in parentheses at 

 the beginning of the description of each species. 

 Morphometries are included only from individuals 

 >50 mm. Fin ray counts are included for all 

 specimens on which counts could be made (Table 

 1). In most cases, rays in both pectoral fins were 

 counted. Fin rays were not counted on specimens 

 with fin damage or on specimens that had rays 

 obscured due to the thick bases of the dorsal and 

 anal fins. Radiography was tried unsuccessfully to 

 obtain vertebral counts: the dermal spines and 

 their bases obscured the vertebrae, and made ac- 

 curate counts impossible. The vertebral counts 



