We took blood samples from 100 post-fledging juvenile Starlings Sturnus vulgaris for DNA sexing in late August, and scored the traits that are commonly thought to predict sex in adults: iris colour, length and shape of the throat feathers, degree of speckling of plumage, body mass and tarsus length. Using logistic regression, the iris colour alone predicted the sex, as determined from the DNA, with 97% accuracy. By using iris colour and the length of the throat feathers together, 98% of birds were sexed accurately and greater separation of groups was achieved. These features were still reliable in the following February, when the birds had come into breeding plumage. At this stage, the colour of the base of the bill was 100% accurate in predicting sex.

Flickering light can cause adverse effects in some humans, as can rhythmic spatial patterns of particular frequencies. We investigated whether birds react to the temporal frequency of standard 100 Hz fluorescent lamps and the spatial frequency of the visual surround in the manner predicted by the human literature, by examining their effects on the preferences, behaviour and plasma corticosterone of European starlings (Sturnus vulgaris). We predicted that high frequency lighting (>30 kHz) and a relatively low spatial frequency on the walls of their cages (0.1 cycle cm −1 ) would be less aversive than low frequency lighting (100 Hz) and a relatively high spatial frequency (2.5 cycle cm −1 ). Birds had strong preferences for both temporal and spatial frequencies. These preferences did not always fit with predictions, although there was evidence that 100 Hz was more stressful than 30 kHz lighting, as birds were less active and basal corticosterone levels were higher under 100 Hz lighting. Our chosen spatial frequencies had no overall significant effect on corticosterone levels. Although there are clearly effects of, and interactions between, the frequency of the light and the visual surround on the behaviour and physiology of birds, the pattern of results is not straightforward.

Most birds have visual sensitivity to ultraviolet (UV) wavelengths, and this sensitivity appears to play a role in their colour vision. Artificial lighting is normally deficient in UV wavelengths. Hence, there may be welfare implications for captive birds kept under such lighting. We investigated whether the absence of UV wavelengths during rearing adversely affects Japanese quail (Coturnix coturnix japonica). We also investigated the short-term effect of switching from UV-containing to UV-deficient lighting, and vice versa. Stress was assessed by monitoring behaviour and plasma corticosterone levels. We did not detect any significant difference in these variables between birds reared either with or without UV. We conclude that rearing quail in an absence of UV does not appear to have a significant impact on their welfare, as measured using these indicators.