542 



Fishery Bulletin 98(3) 



Identiflcation of larvae 



Identification of Lepidopsetta larvae was accomplished by 

 a variety of methods because the more traditional serial 

 approach was not possible. Initially, our unidentified pleu- 

 ronectid series was grouped together on the basis of a 

 continuous sequence of shared characters. This series, how- 

 ever, could not be linked with a recognized pleuronectid 

 taxon whose larvae had not already been described. The 

 unidentified series (Psettichthys 2) was similar to larvae 

 previously described for P. melanostictus. Larvae of Pset- 

 tichthys 2 were determined to be Lepidopsetta by rearing 

 wild-caught larvae' and finally confirmed to be Lepidop- 

 setta by rearing larvae spawned from Lepidopsetta adults 

 collected in Puget Sound (Mulligan et al., 1995). 



Nomenclature of larval developmental stages follows Ken- 

 dall et al. (1984) where each stage is based on widespread, 

 fundamental features of development. The early life his- 

 tory stages used in our study were based on the flexion of 

 the notochord that accompanies the hypochordal develop- 

 ment of the homocercal caudal fin. The onset of the juvenile 

 stage is defined in our study as completion of eye migration 

 and attainment of the adult complement of fin elements. 

 Early juveniles (approximately 20-35 mm SL) are defined 

 as those collected in the water column; many had rem- 

 nants of larval pigmentation on their blind side. Postsettle- 

 ment juveniles were collected with bottom gear and rarely 

 retained any larval pigmentation on their blind side. 



Only melanistic pigment is described in our study 

 because formalin fails to preserve other pigments (live 

 larvae have yellow, orange, and brown chromatophores 

 associated with the melanophores). Previous descriptions 

 of larval pigment patterns, particularly those of pleuro- 

 nectids, have used confusing terminology. Postanal pig- 

 ment patterns have been variously described as vertically 

 oriented bands (Pertseva-Ostroumova, 1961; Moser et al., 

 1984; Matarese et al., 1989) and bars (Charter and Moser 

 1996). The number of postanal bands and bars may or may 

 not include the caudal or terminal notochord pigment. For 

 clarity, in the descriptions of pigment patterns, the terms 

 "band" and "bar" refer to aggregations of melanophores 

 that approximate vertically oriented rectangles (including 



the caudal pigment). A band is always complete and a bar 

 incomplete. A "stripe" approximates a line or elongate rect- 

 angle and is horizontally oriented. A "spot" is an approxi- 

 mately circular aggregation of melanophores. A "patch" is 

 any other distinguishable aggregation of melanophores. 



Larval morphology 



A total of 115 eastern North Pacific Lepidopsetta larvae 

 were measured with a calibrated digital image analysis 

 system (larvae of the western North Pacific L. rnochigarei 

 were not available). This system consists of a video camera 

 attached to a dissecting stereomicroscope or camera lens, 

 a microcomputer with digital imaging board, and a video 

 monitor All measurements of larvae were taken from the 

 left side. Unless otherwise noted, standard length (SL) is 

 used throughout and is defined in our study as the length 

 from snout tip to notochord tip (prior to development of 

 the caudal fin) or to the posterior margin of hypural ele- 

 ments. Other measurements are defined as follows; head 

 length (HL), snout tip to posterior edge of opercle (to pec- 

 toral-fin base in small larvae before opercular margin is 

 visible); snout length (SNL), snout tip to anterior margin 

 of orbit; orbit length (OL), gi-eatest length of orbit; body 

 depth (BD), vertical distance from dorsal to ventral body 

 margin at a vertical line through center of anal opening; 

 snout-to-anus length (SAL), distance along body midline 

 from snout tip to a vertical line through center of anal 

 opening. Standard proportional measurements were cal- 

 culated (Table 2). 



Distributional analysis of larvae 



The dataset used for mapping and calculation of mean 

 density of larvae (number per 10 m-; Appendix Table 1) 

 was a subset of the AFSC ichthyoplankton historical data- 

 base (1972-1994) from the Bermg Sea and Gulf of Alaska. 

 This subset included all reidentified samples that had 

 been originally identified as Psettichthys 2, Lepidopsetta 

 2, and L. bilineata and thus provided the only historic dis- 

 tributional records for lai-vae of L. polyxystra n. sp. Maps 

 of material examined and distribution and mean density 



