To evaluate the possibilities of applying plasma retinol as a biomarker of response in seabirds exposed to chronic low levels of organochlorines, the relationship between yolk content of polychlorinated biphenyls (PCBs) and plasma retinol levels were studied in newly hatched shag chicks (Phalacrocorax aristotelis) from the coast of central Norway. The mean concentration of 29 PCB-congeners (SigmaPCB) in the yolk sac was 1.22 microgram/g ww (wet weight basis) (SD = 0.57, n = 10), or 17.99 ng/g lw (lipid weight basis) (SD = 6.26, n = 10). Expressed as TCDD-equivalents (SigmaTEQ), the exposure in the yolk sac was 43.9 pg/g ww (SD = 19.5, n = 10), or 637.1 pg/g lw (SD = 240.8, n = 10), considerably lower than the levels that have been associated with clear-cut lethal and sublethal effects such as egg mortality, hatchability, or live deformity in Phalacrocoracidae species. There were significant negative correlations between SigmaPCB ww and the variables egg volume, yolk mass, and hatchling mass. We suggest that these relationships are passive causes of a higher lipid concentration in small eggs, rather than the PCB affecting the variables. Analyses showed that there was a borderline significant positive correlation between SigmaPCB lw in yolk and plasma retinol concentration. Although the results indicate that plasma retinol level alone is a poor indicator of PCB exposure in shag hatchlings, the result may be related to the low level of contaminant exposure and the low sample size of the study.