Rodgveller et al.: Effects of maternal age and size on embryonic energy reserves and developmental timing of Sebastes maliger 
39 
Darkening retina 
Otoliths 
Light retinal pigrrient 
Silver pigment 
increased 
Peritoneal 
l wall pigment 
Silver pigment 
Complete silver 
pigment coverage 
icrea^edyellow jL 
'^-definition- 
Lower mandible 
Figure 2 
Photographs of developing quillback rockfish ( Sebastes maliger) embryos and 
preparturition larvae at various developmental stages: (A) stage 7, (B) stage 8, 
(C) stage 9, (D) stage 10, (E) and (F) stage 11, (G) stage 12, and (H) stage 13. 
confounding with stage, within each sampling period, 
by subtracting the expected values obtained from a 
GLM. Adjustment of the maternal factors would not 
be necessary if they were randomly distributed across 
combinations of stages and sampling periods (i.e., if 
there was the same distribution of maternal factors in 
each stage-sampling period cell). Because they were not 
randomly distributed, the confounding between stage 
and the maternal factors necessitates the removal of 
stage effect. The GLM used was 
Ajj = intercept +/ 3 1 S i + fi 2 D / + P 3 (S i x D f )+ e lJt (2) 
where A tJ - the predicted, i.e., the expected age, length, 
or weight for the i th stage and the j th sam- 
pling period; 
