Cyema in the following respects: Cyema has five 

 intestinal loops (four in early stages) which are 

 more compact and deeper whereas this larva has 

 only three shallow intestinal loops; the liver in 

 Cyema is small, laminar, and situated at myo- 

 tome 6 whereas in this larva the liver is a very 

 thick lobe at myotome 17; the pancreas in Cyema 

 is a large and thin film of tissue extending along 

 all the intestinal loops except the last and does 

 not form a bulge with the liver whereas it is a 

 thick lobe forming a bulge with the liver in this 

 larva; the position of the gills in Cyema is more 

 anterior than in this larva; the body depth in this 

 larva is less than that oi Cyema; the pigmenta- 

 tion on the myotomes in Cyema is scattered all 

 over the body whereas it is limited to a series of 

 five midlateral melanophore patches in this larva. 

 But the basic characteristics of the larva are so 

 strikingly similar to those of the larva of Cyema, 

 I am compelled to relate it to an unknown species 

 of the family Cyemidae. If the larva is a cyemid 

 larva, it will probably belong to a new genus 

 other than Cyema as the differences between the 

 larva of Cyema atrum and this larva appear to be 

 at generic level. 



Comparison with saccopharyngid and eury- 

 pharyngid larvae: In all three kinds of lepto- 

 cephali the size and shape are approximately 

 similar, the myotomes are V-shaped, the suspen- 

 sorium is elongated, the gills are small and more 

 posterior in position, the liver is a thick lobe, the 

 pancreas is a thick lobe forming a bulge with the 

 liver, the intestine is looped, and the opisthone- 

 phros and the last blood vessel are on the last 

 intestinal loop. However, this larva differs from 

 saccopharyngid and eurypharyngid larvae in the 

 shape of the head and the nature of the teeth, 

 in having more intestinal loops and a longer 

 intestine, and in the presence of a midlateral 

 series of pigmentation spots. 



Comparison with metamorphic monognathids: 

 This larva resembles the metamorphic forms in 

 myotome shape, the elongate snout, the total 

 number of myotomes and the midlateral melano- 

 phores (Table 2), the structure and position of 

 the melanophores in relation to mytome number, 

 and the pigmentation at the tip of the jaws. 

 But this larva has a well-developed eye whereas 

 the metamorphic forms have rudimentary eyes. 

 But degeneration of the eye may take place during 



FISHERY BULLETIN: VOL. 72. NO. 2 



metamorphosis as in Cyema, which has a 

 degenerate eye in the adult and a very large 

 eye in the larva. The gills, liver, and intestine 

 are lost due to damage in the 42-mm and 48-mm 

 metamorphic specimens, and a comparison of 

 these structures cannot be made. A pectoral fin 

 is absent in metamorphic forms whereas this 

 larva has a pectoral fin. The position of the vent 

 is more anterior in the metamorphic forms than 

 in Monognathus , which may be attributed again 

 to metamorphosis. The deep pigmentation at the 

 base of the median fins, which increases progres- 

 sively in later stages, is obviously juvenile 

 pigmentation. Although the midlateral pigmen- 

 tation, myotome number and shape, and other 

 characters agree with those of metamorphic forms 

 of Monognathus , the differences preclude a close 

 relationship. 



AFFINITIES OF 



SACCOPHARYNGOIDEI WITHIN 



THE ANGUILLIFORMES 



The Saccopharyngiformes have not been suc- 

 cessfully related to any family within the Anguil- 

 liformes. In the most recent classification of the 

 teleostean fishes (Greenwood et al., 1966), the 

 group is placed next to Aoteidae and Cyemidae 

 as a suborder (Saccopharyngoidei) of the order 

 Anguilliformes. The family Cyemidae has been 

 traditionally regarded as related to nemichthyid 

 eels because of the superficial resemblances of the 

 beak. I suggest that the Cyemidae be considered 

 as related to the Saccopharyngiformes and not 

 to the Nemichthyidae for the following reasons. 



The adults of Cyema differ from the nemich- 

 thyids in morphological and osteological char- 

 acters. All the nemichthyid eels are extremely 

 elongate, but Cyema is very short. The adult 

 Cyema has a small degenerate eye and a large 

 stomach (about one-fourth of the total length 

 excluding the beak), as in the Saccopharyngi- 

 formes, whereas the nemichthyids have large 

 eyes. 



The differences in their larvae are even more 

 basic. The larvae of Nemichthys scolopaceus 

 Richardson, 1848 (Bertin, 1937b) and other 

 nemichthyids (Beebe and Crane, 1936, 1937a, 

 1937b) are also elongate and become extremely 

 attenuate during growth and metamorphosis, but 

 the larva of Cyema has a short and deep body. 



560 



