‘Times Species-counts’ are a simple method of comparing the avifaunas of extensive areas by sampling representative habitats. The method has been evaluated in south central Kenya but could be applied to most terrestrial regions of tropical Africa. The essential feature is that the observers move anywhere within a sample site of about 1 km2, recording all of the species seen in a fixed time, e.g. 1 h.

Counts are made in all seasons and at various times of day. Reasonable estimates of the total number of species in an area can be derived from a series of 15 counts, but such estimates tend to be low in non-forested habitats because ‘wanderers’ (non-resident species) continue to be recorded almost indefinitely. Open sites have far fewer species than well-wooded areas. The numbers of resident species in non-forested terrestrial habitats is related both to the amount of woody vegetation and to a measure of plant growth. However, total species numbers seem to depend solely on the amount of woody vegetation. Road counts, from a vehicle, were compared to those made on foot. They are less satisfactory, especially for cryptic species.

Otter spraints from five Danish freshwater localities were analysed. In all localities fish was the main prey (76–99% of estimated bulk), especially in winter. Depending on locality, the prey fish mainly consisted of cyprinids (Cyprinidae), percids (Percidae) or salmonids (Salmonidae). Seasonal variations were found in the diet corresponding to the availability of prey. Non-fish prey categories, such as frogs, birds, mammals and invertebrates, were most frequently taken in spring and summer, but only frogs made an important contribution to the diet (0–21%). The results of analysing 978 otter spraints were compared with prey fish populations as estimated by electrofishing. It was concluded that the fish species composition in the otter diet generally reflected that of the foraging area, however, with the exception of a negative preference for trout (Salmo sp.) and a preference for sticklebacks (Gasterosteidae) in some of the study areas. Otters generally took all available sizes of fish with a tendency for selecting individuals between 9–21 cm, though the preferred size of eels was 18–36 cm. A principal component analysis showed that the variation in diet between the localities was associated with differences in fish communities between free-flowing rivers and lake outflows.