This study investigated the feasibility of using clove oil as an alternative to tricaine methanesulphonate as a ®sh anaesthetic, particularly in ®sh stress research. The physiological stress responses of juvenile chinook salmon Oncorhynchus tshawytscha (Walbaum) anaesthetized with either tricaine (50 mg L ±1 ) or clove oil (20 p.p.m.) were compared using unanaesthetized ®sh as controls. Haematocrit, serum cortisol and serum glucose concentrations, serum lysozyme activity and differential leucocyte counts were measured from blood samples collected before, during and upon recovery from anaesthesia and at speci®ed intervals up to 72 h after recovery. Differences between the two anaesthetic groups were not signi®cant for most of the physiological traits measured. Serum lysozyme activity of control ®sh, however, was signi®cantly suppressed relative to the treated ®sh for 72 h after stress. Clove oil may be a safe and cost-effective alternative to tricaine without signi®cantly affecting study results. Furthermore, clove oil may be more practical for ®eld-based research, because a withdrawal period is unnecessary, and clove oil does not pose an environmental hazard.

Induced triploidy (3N) in salmon results from a blockage of maternal meiosis II, and hence provides a unique opportunity to study dosage effects on phenotypic variance. Chinook salmon families were bred using a paternal half-sib breeding design (62 females and 31 males) and half of each resulting family was treated to induce triploidy. The paired families were used to test for dosage effects (resulting from triploidy) on (1) the distribution and magnitude of phenotypic variation, (2) narrow-sense heritability and (3) maternal effects in fitnessrelated traits (i.e., survival, size-at-age, relative growth rate and serum lysozyme activity). Quantitative genetic analyses were performed separately for diploid and triploid family groups. Triploidization resulted in significantly higher levels of phenotypic variance and substantial differences in patterns of variance distribution for growth and survival-related traits, although the patterns were reversed for lysozyme activity. Triploids exhibited higher narrow sense heritability values relative to diploid Chinook salmon. However, maternal effects estimates were generally lower in triploids than in diploids. Thus, the dosage effects resulting from adding an extra set of chromosomes to the Chinook salmon genome are primarily additive. Somewhat counterintuitively, however, the relative magnitude of the combined effects of dominance, epistasis and maternal effects is not affected by dosage. Our results indicate that inheritance of fitness-related quantitative traits is profoundly affected by dosage effects associated with induced triploidy, and that triploidization can result in unpredictable performance and fitness outcomes.