Road traffic affects the natural environment in numerous ways. The most striking of these is the death of wild animals and birds as a result of collisions with moving vehicles. In this paper the available data on bird mortality on roads are reviewed. Estimates of annual mortality for some European countries (350 000 to 27 million birds), the monthly distribution of casualties, their distribution among sex and age classes, as well as the methods used in the study of this problem are presented. The species composition of birds killed in this way is compared for several countries. In western Europe sparrows and Blackbirds are the species that most frequently die on the roads, but in Central and Eastern Europe not only sparrows but also corvids and Barn Swallows make up a high proportion of the victims. Analysis of the monthly distribution of casualties in 10 species shows this to differ between countries, probably because of the geographic variation of certain aspects of their biology (migration, breeding etc.). Several factors affecting the frequency of casualties are discussed, and some suggestions for the prevention of bird casualties are also given.

The natural tree holes and nest holes of hole-nesting birds were surveyed in four forest types in the west Khentii Mountains of NE Mongolia. The utilization patterns of species, size and condition of trees, as well as hole types, were investigated. The average density of tree holes in the study area approached 30 holes/ha, while that of hole-nesting birds was 2.4 nests/ha only. The riparian mixed forest had the greatest number of species and individuals of hole-nesting birds, while the spruce-fir forest had the lowest numbers. Excavating bird species preferred larger, deciduous trees, and snags. Non-excavators did not select holes according to tree species or size, but preferred holes in living trees and branch holes. In view of the low occupancy of holes among the four habitats, we suggest that the density of secondary hole-nesting birds is not limited by availability of holes in the study area.

The purpose of the research was to find a time frame for the beginning, duration of maximum intensity, and end of nocturnal bird migration. The research was carried out using photo registration on a radar screen during the spring and autumn seasons of 1999–2001. Examination of the average length of daylight at the beginning of migration and the onset of civil twilight yielded a high correlation factor for both spring and autumn. The results showed that, on average, nocturnal migration began at the onset of civil twilight, that is, half an hour after sunset. The time elapsing between the onset of migration and the maximum concentration of migratory birds averaged about 70 minutes in both spring and autumn. Nocturnal migration usually came to an end within the one and a half to two hours after sunrise. We ascertained the seasonal time shift for the onset of nocturnal migration corresponding to the seasonal time shift related to the approach of darkness. The average times of the beginning, maximum intensity and end of nocturnal migration were found to be related to photoperiodic factors.

The study was carried out from May 1988 to April 1993 in two types of tropical forests: the tropical evergreen and moist deciduous. The vertical distribution of birds and foliage abundance was recorded by visual observation in seven height classes, using the line transect method. The vertical distribution of 94 bird species was recorded in evergreen forest, and of 90 in moist deciduous forest. Bird species richness and the numbers of birds were highest in the stratum from 1 to 5 m in both the evergreen forests (57 species) and the moist deciduous forests (68 species). The species richness indices of birds in the two types of forest did not display such a clear relationship. Foliage abundance was high up to a height of 30 m in the evergreen forests, whereas in moist deciduous forests, foliage was abundant only up to 20 m. A significant positive correlation was obtained between foliage abundance and species richness and also between the numbers of birds in both types of forests. Significant correlation was also obtained between foliage abundance and the species diversity indices in the two types of vegetation.

The seasonal pattern of singing activity was studied in the Poznań region, W Poland. Males sang from mid-March till late July with a song rate varying between 3.2 and 11.1 songs/min. Song activity peaked at the beginning of the breeding season, in late March and April. In following months males also sang at quite a high rate, but the probability of finding a singing male within the territory was lower. Males sang mainly from tree tops and power lines at a height between 8 and 10 m. Males with larger numbers of neighbours had a significantly lower mean song rate than those with a single neighbour. Most probably, males with more neighbours had to spend more time on more active and direct territorial defence, and/or listening to the song of other males. The pattern we found suggests that song is mainly used in territorial defence against rivals, and that it is used as a first line of defence. Nevertheless, this does not preclude its usefulness for attracting a mate.

The variation in size and shape of Bearded Tit eggs was investigated in the Wielkopolska Region of western Poland in 1988–1992 and 1997–2000. The mean clutch size was 5.47 (95% CL: 5.25–5.70, n = 99), and differed markedly from year to year. Coefficients of variations for the mean egg characteristics in a clutch ranged from 1.91 (breadth) to 4.90 (volume). No significant correlation between egg length and breadth was found. Repeatability estimates were 0.50, 0.48, 0.50, 0.47 for length, breadth, volume and elongation index, respectively. The results suggest a relatively low heritability of egg dimensions in the population studied.

We studied the foraging behaviour of Lesser Kestrels in agricultural habitats during the breeding season of 2000. The birds spent more time hunting in flight than perched. During 398 min. of observed aerial hunting, they spent 23.7% hovering, 14.4% hanging, 14.0% flapping, 41.2% gliding, and 6.7% soaring. The time spent on each type of aerial hunting behaviour depended on factors like breeding stage, time of day, wind speed, number of strikes, number of successful strikes, and the time spent hunting. The strike rate was 0.38 per min., the capture rate 0.10 per min. The capture rate depended on the type of hunting behaviour preceding the attack and was highest after the birds had been hovering.

We monitored nest sites of endangered Interior Least Terns on a 5 095 ha alkaline flat in north-central Oklahoma, USA. After nest loss, Least Terns commonly renested and experienced 30% apparent nest success in 1995–1996 (n = 233 nests). Nest success and predation differed by location on the alkaline flat in 1995 and overall, but nest success and flooding did not differ by microhabitat type. Predation was highest at nests ≤ cm from debris (driftwood/hay) in 1995. No differences in nesting success, flooding, or predation were observed on comparing nests inside and outside electrified enclosures. Coyotes and Striped Skunks were confirmed nest predators, and Ring-billed Gulls were suspected nest predators. We identified one location on the alkaline flat of about 1 000 ha with consistently lower nest losses attributable to flooding and predation and the highest hatching success compared with other parts of the alkaline flat; it was typified by open ground and bisected by several creeks. Management activities that minimize flooding and predation in this area could further enhance nest success and theoretically increase overall productivity of this population of Least Terns. However, the efficacy of electrified enclosures and nest-site enhancements, as currently undertaken, is questionable because of considerable annual variation in use by and protection of Least Terns.

The study area (16 km²) in “Ujście Warty” National Park, W Poland — was the valley of a lowland river at its confluence with the River Odra, covered by a mosaic of grassy vegetation and willow scrub. 111 breeding attempts were recorded during 2000–2002. The mean nest density (3.2 nests/km²) was higher than that recorded by other authors in agricultural landscapes, but lower than in urban areas. The nest construction was adapted to fit young willow trees. The mean clutch size was similar to that recorded in other populations (4.43), but eggs were smaller (41.2 mm × 29.1 mm). The hatching success was lower (76%) in comparison with other studies, but the mean number of fledglings (2.15 per nest and 2.96 per nest in successful broods) was relatively high. The main reasons for losses were unhatched eggs, predators, starved nestlings and poor nest construction. We hypothesise that the smaller egg size and lower hatching success recorded in this population was due to unfavourable and unpredictable feeding conditions (floods) during the period of egg formation and egg laying. Later in the season, receding floodwaters laid bare areas suitable for foraging on invertebrates; waterfowl eggs also became readily available. Predation was low (lack of nonbreeding stock of Hooded Crow). As a result of good conditions during chick rearing, the overall reproductive output was relatively high in comparison with other populations.

In this study, nest characteristics (size and proportions of basic components) were not correlated with the timing of breeding. Clutch size was negatively correlated with total nest mass but positively correlated with the proportion of the mass of the lining in the total nest mass. Analyses of hatching and fledging success showed that the quantity and proportion of moss in the nest structure as well as the nest size influenced the performance of eggs and nestlings at the nest. We suggest that variation in nest size and composition may be due to several contradictory pressures associated with the need to keep the moisture and temperature in the nest relatively constant, to protect the brood from predation, and to control sanitary standards.

The diet of chicks was investigated in an intensively managed farmland in South Africa, after a prolonged drought, during the second brood (February/March). In comparison with other African studies, a larger proportion of frogs (mainly juvenile Pyxicephalus adspersus frogs) and a smaller proportion of grasshoppers (locusts and crickets) was recorded. This may have been due partly to exceptionally low rainfall rather than to different habitat or geographical location. The large proportion of locusts and flies in the diet indicates that the Cattle Egret may play an important role in pest control.