17The aim of this empirical study was to investigate whether aspects of the care and conditions of 18 harbour seal (Phoca vitulina) pups in rehabilitation centres may be reflected in urinary concentrations 19 of glucocorticoid hormones (GCs). Urine samples were collected non-invasively from pups at five 20 different rehabilitation centres, three on the Irish Sea and two in the southern North Sea.
21Concentrations of urinary cortisol (CL) were analysed by ELISA from one centre and subsequently 22 concentrations of CL, cortisone (CN), prednisolone (PL) and prednisone (PN) were analysed by mass 23 spectrometry. Urinary concentrations of PN and PL were found to be of similar magnitude to CL. CL,
24CN and PL concentrations were reciprocally correlated, but PN was not correlated with CL and only 25 weakly correlated with the other three GCs. The strongest associations between GC concentrations and 26 pup conditions and procedures were negative correlations of PN with growth rate of individual pups 27 and body weight. Differences in GC concentrations, particularly CL, between pups in different social 28 groupings (pups alone, paired or in groups of more than two) were also detected. On the basis of our 29 results, we suggest that harbour seal pups should be fed a dietfrom first entry to a rescue centre -30 which maximises weight gain; this should not only help to prevent chronically elevated GC levels 31 during the rehabilitation period, but would also minimise time spent in captivity.32 33