464 



Fishery Bulletin 90(3). 1992 



Postflexion PAVM pigment takes the form of dashes 

 at the base of each anal ray. At least 3 melanophores 

 can be found on the head over the midbrain. In larger 

 postflexion larvae, the number of PAVM is greatly 

 reduced (Fig. 7). 



Small juveniles (13.2 mm, LACM W67-153, not illus- 

 trated) have numerous melanophores over the mid- 

 brain. Few or no PAVM may be present (Fig. 7). In 

 the 13.2mm juvenile, a dark patch of melanophores on 

 the pectoral fin base extends to and around the pelvic 

 girdle and meets at the ventral midline. Bands of pig- 

 ment extend down from the dorsum and stop just ven- 

 tral to lateral line. 



In larger juveniles (Fig. 6D) a patch of melanophores 

 is present on the pectoral fin base, but may not be con- 

 tinuous across the pelvic girdle as it is in the 13.2 mm 

 juvenile. Light circles appear in the dense pigment 

 below the lateral line. 



Meristics Meristics for 0. triads (Table 3) are com- 

 parable to published accounts. Modes for the fin ele- 

 ments matched those given in Howe and Richardson 

 (1978). Vertebrae (35) were 1 greater than the mode 

 (34) in Howe and Richardson. Branchiostegal rays (BR) 

 form during flexion; a 5.6mmFL larva had 5 visible 

 BR. In postlarvae and juveniles, branchiostegal rays 

 = 6, PCR = 6-h6. 



Comparison with otiner species 



Orthonopias triads are similar to Artedius meanyi 

 larvae (Washington 1986) by possession of 4 preoper- 

 cular spines, a short compact gut, and an eye cirrus; 

 A. meanyi postflexion larvae and juveniles also develop 

 small, prickly scales on the head and below the dorsal 

 fin. Artedius meanyi differ in having far fewer PAVM 

 (<13), pigment in the dorsal finfold, anterior gut 

 melanophores, and in undergoing flexion at a larger 

 size (6.2-9.4 mm). Artedius meanyi and 0. triads were 

 put in the "Myoxocephalus" group by Washington et 

 al. (1984) due to the presence of 4 preopercular spines. 



Orthonopias triads larvae are similar to others 

 within the "Myoxocephalus" group, including Icelinus 

 and Chitonotus, in having no heavy nape pigment and 

 a high number of PAVM; Icelinus quadriseriatus has 

 25-63 PAVM (Feeney 1987) and Chitonotus has 24-45 

 PAVM (Richardson and Washington 1980). Orthono- 

 pias triads lacks ventral gut pigment (see Appendix 1). 



Orthonopias triads can not be assigned to the "Ar- 

 tedius/Clinocottus/Oligocottus" group, as tentatively 

 suggested by Richardson (1981), because it lacks the 

 multiple preopercular spine pattern, gut diverticulae, 

 and trailing gut. Clinocottus analis postflexion larvae 

 (this paper) are similar to 0. triads because of the 

 advanced anus and presence of head pigment, nasal 



> 



Figure 7 



Frequency of postanal ventral melanophores (PA\'TM) vs. 

 length (mm) with a LOWESS regression line at F = 0.5 (half 

 the points included in a running window) for Orthonopias 

 triads larvae and juveniles. 



spines, post-temporal/supracleithral spines, cirri over 

 the eye, and similar meristics. Clinocottus analis dif- 

 fers in having multiple (>5) preopercular spines, a 'W' 

 shaped pigment patch on the side of the body, and a 

 longer gut (preanal = 46.0-54.5% SL vs. 39.i-48.3% 

 SL in 0. triads). 



Orthonopias triads larvae initially have more PAVM 

 than C. analis; however, the number of PAVM de- 

 creases with length more quickly than C. analis (Figs. 

 6, 7); linear regressions (not shown) of 0. triads 

 PAVM have a greater negative slope (-2.413 vs. 

 - 1.161) than C. analis. Linear regression lines were 

 not used, however, in the final plots (Figs. 6, 7) because 

 LOWESS smoothing (Wilkinson 1989) indicates that 

 the relationship between PAVM and length may be 

 nonlinear, especially in 0. triacis. Additional large 

 postflexion and juvenile specimens need to be exam- 

 ined to verify this relationship. 



Occurrence 



During 1978-84, 0. triacis larvae (like C analis) were 

 collected in highest densities off Palos Verdes and other 

 rocky areas, at the 15 m isobath during the entire year, 

 peaking in spring and summer. Approximately 72% of 

 the larvae in discrete depth tows were collected in 

 epibenthic tows and none in neuston tows. Flexion lar- 

 vae were rarely collected. Postflexion individuals have 

 not been found in the 1978-79 neuston tows as were 



