Otolith micro-increment formation in herring Clupea harengus larvae in relation to growth rate

Abstract

Estimating the age of individual fish larvae using otolith micro-increments has become a widely used tool in early life-history ecology. A basic assumption of the method is that the micro- increments are formed on a regular (usually daily) basis. Validation experiments are recommended to test this assumption, since otolith microstructure varies between species and interpretation of the structures can be problematic. Whilst many such experiments have demonstrated daily deposition rates, those examining slow-growing larvae have often not supported this assumption. In response, it has been suggested that increments are in fact being formed on a daily basis, but that they are too narrow to be resolved by optical microscopy. In this study, we raised herring larvae over a range of growth rates by manipulating prey levels. At 16 and 37 d old, the otoliths were marked by immersion of the larvae in a solution of alizarin complexone. The rearing experiment was terminated at 51 d post-hatch, and after suitable preparation, otoliths were examined using both light and scanning electron microscopy. Micro-increment counts based on light microscopy were lower than expected (assuming daily deposition) when larval growth rates were less than 0.42 mm d -1 . These results are in accord with previous reports in the literature. Examination of the otolith region between the alizarin marks by scanning electron microscopy (SEM) produced slightly higher increment counts compared with light microscopy (average + 2), but this was not sufficient to support the assumption of daily deposition of micro-increments between Days 16 and 37 in slow-growing larvae. Rather than leading to the regular deposition of narrow increments, slow growth in laboratory-reared herring larvae often appears to affect otolith structure. This results in fewer than the expected number of increments being deposited, and affects counts made both by light and scanning electron microscopy. Growth rates attained in this laboratory study are typical of those that may be experi- enced by cohorts of herring larvae in the wild, particularly from autumn spawning stocks. The use of otolith microstructure to estimate absolute ages of slow-growing herring larvae may therefore lead to significant underestimates of true age, even if SEM is used.