Markle et al,: Metamorphosis of Microstomus pacificus 



299 



influencing an area up to 500 km off the shelf (McLain 

 and Thomas 1983). Its velocity is < 10 cm/second over 

 the continental slope north of Cape Mendocino (Hickey 

 1979) and weaker seaward of the slope. Somewhat fur- 

 ther south, between Pt. Arena and Pt. Reyes, the cur- 

 rent is 3-10 cm/second from July to October and <1 

 cm/second from October to January (Huyer et al. 1989). 

 If eggs and Stage-1 larvae are displaced, on average, 

 555km (260-1-295) southward of their spawning site, 

 a northward-flowing undercurrent of 3.25cm/second 

 would be sufficient to return Stage-2 larvae to the 

 vicinity of their spawning site in 6 months. This does 

 not seem to be an unreasonable average velocity for 

 the undercurrent from July to January. 



The depth range of the Stage 3 "landing" zone 

 (55-377 m) corresponds with the northward under- 

 current located at 200-300 m (Huyer and Smith 1985). 

 However, these larvae appear to need a mechanism to 

 bring them shoreward. The surface Ekman layer, 

 0-20 m, within which wind-driven transport occurs 

 (Huyer 1983), could be reached if larvae moved up in 

 the water column during storms. Dial offbottom migra- 

 tions could be part of this mechanism. Alternatively, 

 as the body surface area is reduced during this stage, 

 larvae may become less passive and move actively 

 inshore. 



Delayed metamorphosis and settlement 



The protracted process of metamorphosis in Dover sole 

 is contrary to expectations based on the ideas of salta- 

 tory ontogeny (Balon 1981). In general, ontogenetic 

 transformations are expected to occur rapidly because 

 intermediate forms are presumed to be maladapted. 

 For example, loss of teeth from the right side of the 

 jaw and development of incisors on the left side seem 

 to hold no advantage for a planktonic larva, yet this 

 is the situation in Dover sole for several months dur- 

 ing the precompetent Stage 2. Delayed metamorphosis 

 is also related to the concept of saltatory ontogeny; 

 because the transition is assumed to be quick, an 

 organism without the proper cues simply delays meta- 

 morphosis and settlement. In other words, it keeps the 

 morphology appropriate for the habitat. Typically, field 

 researchers identify a minimum threshold size or 

 developmental stage for metamorphosis and assume 

 that planktonic specimens greater than the threshold 

 size or in the threshold stage have delayed metamor- 

 phosis (Pechenik 1986). Others have used a minimum 

 age as a threshold (Cowen 1991). 



Pearcy et al. (1977a) suggested that larger "hold- 

 over" Dover sole larvae (>50mmSL) delayed meta- 

 morphosis and few successfully recruited to the benthic 

 juvenile stage. Delayed metamorphosis is predicted 

 for coastal organisms subjected to offshore transport 



(Jackson and Strathmann 1981) and there is some 

 evidence for delayed metamorphosis in fishes (Victor 

 1986, Cowen 1991). An advantage of delayed metamor- 

 phosis is extension of the settlement season beyond 

 what would be expected based on the spawning season 

 (Victor 1986). Contrary to this expectation, the dura- 

 tion of Dover sole settlement is seasonally restricted 

 and, off Oregon, no greater than the duration of the 

 spawning season. Because precompetent larvae are 

 probably a great distance from their settlement site, 

 cues for metamorphosis are likely to be seasonal rather 

 than site-related. 



Experimental studies focusing on flounders have 

 showTi (1) fast-growing individuals metamorphose at 

 smaller sizes, (2) fast-growing individuals retain their 

 faster growth rate for at least several weeks after 

 metamorphosis, (3) age at metamorphosis (defined by 

 eye migration) is more variable than size at metamor- 

 phosis, and (4) a target size or threshold must be 

 reached prior to metamorphosis (Policansky 1982, 

 Chambers and Leggett 1987, Chambers et al. 1988). 

 Other fishes and organisms may have age-triggered, 

 size-triggered, or age- and size-triggered metamor- 

 phosis (Policansky 1983). Policansky (1983) points out 

 that a size threshold would be expected when there is 

 a size difference in available food between different 

 habitats or a minimum energy requirement to success- 

 fully function at a certain stage. 



We suggest two contrasting interpretations of the 

 early life history of Dover sole. If size and age at 

 metamorphosis are positively correlated, as is the case 

 in winter flounder (Chambers et al. 1988), then larger, 

 earlier settlers are older and slower-growing than 

 smaller, later settlers. The difference in age could be 

 the difference between early and late spawners or be- 

 tween different years of spawning. Alternatively, 

 variation in size at metamorphosis may simply reflect 

 differential growth rates operating for a long time, 

 probably at least 2 years. As a consequence, larger, 

 earlier settlers would be the faster growers rather than 

 slower growers. One could distinguish between these 

 alternatives and demonstrate delayed metamorphosis 

 by documenting different year-classes among settlers. 



In terms of life-history strategies, delayed metamor- 

 phosis and protracted metamorphosis may confer 

 similar advantages. Extension of settlement through 

 delayed metamorphosis allows for adaptive responses 

 to short-term oceanographic variability and avoidance 

 of settling during unfavorable conditions. If metamor- 

 phosis and settlement are cued to favorable seasons, 

 protracted metamorphosis and the ability of competent 

 metamorphosing individuals (Stage-3 larvae) to spend 

 several months moving between midwater and bottom 

 habitats should also compensate for any short-term im- 

 favorable oceanographic conditions. 



