Do Otoliths record changes in somatic growth rate? Conflicting evidence form a laboratory and field study of a temperate Reef Fish, Parika scaber

Abstract

Otolith increment widths in a temperate reef fish, Parika scaber (Pisces:Monacanthidae), have been shown to vary in different months and to correlate with changes in somatic growth rate. This results in the persistence of a constant otolith-size/fish-size relationship. In the present study, the response of otolith growth to laboratory-induced changes in body growth was determined over a 10-day period in December 1985. Juveniles were subjected to two ranges of temperature (19-21�C and 24-26�C) and three feeding levels (fully fed, partially fed, and starved) of a diet consisting of larvae of Opifex fuscus, an endemic mosquito. Fish were individually tagged to ensure that changes in growth rate and any associated changes in otolith microstructure could be accurately determined for any individual. Fully fed fish grew more slowly during treatment than they did prior to treatment, and starved fish did not grow at all. Partially fed fish exhibited a size-hierarchy effect, with larger fish increasing in body size at the expense of smaller individuals. No temperature effect on growth rate was recorded. In contrast to the results from an associated field study, otoliths of fully fed fish continued to increase at the same daily width despite the induced alteration in growth rate. Thus, otoliths were disproportionately large compared with estimates from the field. However, otoliths of starved fish were not as large as predicted from a continual production of daily rings at a constant width. Daily ring deposition may have ceased at different times during the experiment for the starved fish, depending on body size. Alternatively, deposition may have continued, but at a spacing that was not detectable by light microscopy. We conclude that adequate validation of response of otolith growth to changes in somatic growth is necessary before individual growth histories can be back-calculated. This is true even where strong evidence from the field suggests that back-calculation is possible.