MARLIAVE and PEDEN: LARVAE OF LIPARIS FUCENSIS AND L. CALLYODON 



a size of 8.8 mm NL, the hypurals were formed 

 without caudal ray anlagen, and the pectoral fin 

 ray bases were present. The anterior lobe of the 

 dorsal fin (fii-st dorsal) was becoming evident. 

 The pelvic disk was only slightly smaller in diam- 

 eter than the eye. At sizes between 9 and 9.9 mm 

 NL, caudal rays formed, numbering 8 or 10 in 

 different specimens, and pectoral rays developed 

 from V2 to over % of their length. All larvae 

 under 10 mm NL were prefle.xion larvae. Larvae 

 between 8.8 and 9.9 mm NL had the same mor- 

 phometries as smaller preflexion larvae, except a 

 reduction from 9 to 7% NL in eye diameter. 



Flexion occuired at sizes just beyond 10 mm 

 NL, together with completion of caudal and pec- 

 toral fin ray formation. At 10.45 mm NL, the 

 exserted lower pectoral rays were elongated, 

 and the anterior lobe of the dorsal fin was becom- 

 ing prominent. Disk width equalled eye width. 

 Caudal pigment was forming, and the melanin on 

 the rest of the body had become denser, al- 

 though the pectoral fin remained unpigmented. 

 The mandibular area was lined with 10 acoustico- 

 lateralis pores (5 each side), and the maxillary 

 region, with 8 pores. The nostrils were spht into 

 separate canals, with the ventral naris du-ected 

 anteriorly in line with the eye pupil (about 66% 

 pupil diameter) and with the dorsal naris di- 

 rected up at a 45° angle to the level of the 

 notochord (the opening half the diameter of the 

 ventral naris). The cranial subdermal space was 

 evident beneath the dense melanin. The oper- 

 culum remained entirely open along the pectoral 

 fin base and did not extend dorsally beyond the 

 pectoral fin. 



During postfiexion (Fig. 3c), ambivalence be- 

 tween swimming and settlement became evident 

 in the rearing tank (n = S individuals). Post- 

 flexion juveniles of about 12 mm NL had formed 

 an acoustico-lateralis pore posterior to the eye, 

 and two pores of the lateral Une, immediately 

 posterior to the dorsal end of the operculum. The 

 operculum extended dorsally beyond the pec- 

 toral base, curving anteriorly as in adults. The 

 opercular membranes with branchiostegals were 

 fused with the isthmus up to about 7 rays from 

 the top of the pectoral fin. 



Juveniles of 20-22 mm SL showed different 

 morphometries from larvae: the snout-anus 

 length was reduced from 42 to 36% SL; the eye 

 diameter was further reduced to 5.7% SL; the 

 head length increased from 21 to 24.4% SL; and 

 the body depth slightly increased to 23.5% SL. 

 In permanently settled juveniles at the largest 

 sizes preserved, the opercular opening remained 



as low as the 4th or 5th pectoral ray; whereas, 

 adult L. callyodon would not have this opening 

 extending beyond the 1st ray (Hart 1973). The 

 only other possible identifications from meristics 

 were L. cyclopus and L. florae, but morpho- 

 metries, fin shape, and ecological information 

 dictated against such determinations. 



DISCUSSION 



It must be reiterated that positive identifica- 

 tion was, and now remains, possible only from 

 juvenile material reared from known larvae. 

 Future work may possibly permit identification 

 from larval material for these Pacific Liparis 

 species, but the present work serves only to 

 focus such future efforts. 



Taxonomic identification of Liparis species de- 

 pends on numerous characters that are very dif- 

 ficult to determine in the smallest juvenile speci- 

 mens. In addition, the opercular opening be- 

 comes smaller with development in Liparis 

 species (Able et al. 1984). Therefore, rigid ad- 

 herance to determining a restricted opercular 

 opening size for identification of L. callyodon (cf. 

 Hart 1973) should be expected to cause difficulty 

 in identification of small juveniles. Further in- 

 vestigation of allometric reduction in opercular 

 opening size in juveniles of the genus Liparis is 

 required. 



The fecundity (fewer than 400 eggs per mass) 

 of L. callyodon was substantially less than the 

 fecundity (about 1,500 to 5,000 eggs) determined 

 for L.fuceusis by DeMartini (1978). The average 

 egg size, however, was larger at 1.7 mm for L. 

 callyodon than for L.fucensis at 1 mm (Marliave 

 1975; DeMartini 1978). Adults of L. fucensis at- 

 tain 30% greater maximum length than L. call- 

 yodon (Hart 1973). Furthermore, the L. cally- 

 odon egg masses collected, especially the counts 

 of 132 and 53, may not have comprised the entire 

 ovarian output of a female in every case because 

 eggs were extruded into available interstitial 

 spaces. This could sometimes restrict the num- 

 ber of eggs laid in one mass. 



The larger eggs and apparent lower fecundity 

 of L. callyodon resulted in considerably larger 

 larvae at hatching than those for L. fucensis. 

 Growth and development of L. callyodon oc- 

 curred over a smaller range of lengths. Con- 

 sidering the cubic increase in volume with length 

 for the bubblemorph of L. fucensis larvae, that 

 unusual morph permitted enormous growth dur- 

 ing the larval stage compared to the more typical 

 Liparis larvalmorph of L. callyodon. Larvae of 



741 



