1. Physiological parameters can give valuable indications about the performance of animals in their environment and the combined deleterious effects of several factors on the individual. The concentration of circulating glucocorticoids is often taken as a measurement of the level of stress an animal is exposed to. As an alternative, the ratio of heterophils (or neutrophils in mammals) to lymphocytes in blood (H/L-ratio) has been proposed. Although laboratory studies with exogenous corticosterone treatment or strong stressors found a positive correlation between circulating corticosterone and the H/L-ratio, studies in free-living birds did not.
2. In Eurasian kestrel nestlings Falco tinnunculus, we first confirmed that implanting exogenous corticosterone elevated the H/L-ratio and depressed the immune response to PHA in altricial nestlings under field conditions, as found in precocial species.
3. In non-treated, free-living kestrel nestlings we showed that circulating corticosterone concentration and the H/L-ratio were not correlated among each other and that various environmental factors were not correlated with baseline corticosterone concentration to the same degree as with the H/L-ratio. Baseline corticosterone concentrations were only elevated when nestlings had no body fat stores and as a reaction to our presence at the nest. In contrast, the H/L-ratio was increased as a response to five potential natural stressors: being later-hatched and thus competitively inferior, low body fat stores, ectoparasite infestation, after rain which impaired food delivery of adults and with ongoing season which reflects deteriorating environmental conditions. Baseline corticosterone on day 10 of age was a better predictor of survival until day 13 than the H/L-ratio.
4. In accordance with findings from the literature, it appears that baseline corticosterone concentration and H/L-ratio differ in sensitivity to various stressors. While the H/L-ratio increases already in response to more subtle stressors, both measures react to severe, life-threatening stressors, like starvation in free-living birds, although in our study corticosterone was the better predictor of survival.
5. This study showed that baseline corticosterone and H/L-ratio cannot be used interchangeably as indicators of stress, but together can provide a comprehensive picture about the stress-status of a bird in ecological studies. If only one measure is to be taken, the choice depends on the questions to be answered, practical issues (speed of blood sampling after first disturbance) and physiological interpretability and amenability to experiments.