In recent years, fish models have become popular for investigations of ototoxic agents. However, the vast majority of such studies have focused on anatomical changes in lateral line hair cells after drug administration. Using the goldfish (Carassius auratus), we confirm that the acquisition of auditory evoked potentials offers a rapid and non-invasive method for quantifying ototoxin-induced changes in hearing sensitivity. Gentamicin (100 mg ml(-1)) was the drug of choice as it is a well-studied human ototoxin. Auditory threshold elevation was observed between 300 and 600 Hz and was accompanied by significant reductions in hair cell ciliary bundle densities in specific regions of the utricle and saccule. The correlations between structure and function suggest that differential susceptibility of sensory hair cells to acute gentamicin treatment underlies the frequency-specific elevation of auditory thresholds. We propose that fish auditory systems should be used alongside the lateral line, for the assessment of ototoxicity in new-developed drugs.

Determining and understanding patterns in the reproductive status of fishes are essential for assessing individual and population reproductive potential. The least biased method for determining reproductive stage is through the use of ovarian histology; however, this method can be time consuming and expensive. To overcome these restrictions, we developed models to predict the reproductive stage of female striped bass Morone saxatilis by using ovary color and other female characteristics. We provide a standardized and calibrated method for quantifying ovary color, and we outline an analytical approach that utilizes binary and ordinal logistic regression. Our results indicated that binary models had high accuracy (<6% error) in distinguishing between the regressing phase (i.e., recently spawned) and the nonregressing phase (i.e., all other reproductive phases); we found even better accuracy (4% error) for ordinal models in discriminating among four nonregressing reproductive phases. All of the best‐fitting models required the inclusion of an ovary color variable and ovary energy density; however, good predictive accuracy was also obtained when we replaced ovary energy density with ovary percent water to produce models requiring minimal laboratory processing time and cost. Although tested on striped bass, our method could be used to develop similar models for other species.

In this study, saccular afferent arborization patterns in Atlantic croaker Micropogonias undulatus, red drum Sciaenops ocellatus and spot Leiostomus xanthurus were characterized. Leiostomus xanthurus showed the simplest configuration while M. undulatus displayed the most complex. In addition, hair-cell densities at sites sampled along the rostro-caudal axis of the saccular epithelia correlated with the observed patterns of arborization.