cryotubes at −80 °C for later examination. From images we measured
individual standard and total length using the software Image
Pro 6.2. The frozen larvae were rinsed in distilled water, dried at 60 °C
over 24 h and weighed to estimate dry weight (see protocol in
Seljeset et al., 2010). Size distributions were tested for normality using
the Shapiro–Wilk test. Cumulative size distributions and daily specific
length and dry weight growth rates were estimated as described in
Folkvord et al. (2009). Pairwise comparisons using t tests with pooled
SD were applied to compare larval size distributions. A Bonferroni correction
was applied to the estimates of the probability to avoid type I
error when comparing the size distributions from the three treatments
within experiments. The percentages of survivalwere root-squared and
arcsine transformed for normality prior to statistical analysis. All statistical
analyses were fitted using the R statistical software (Development
Core Team, 2011).