Avian trans-Saharan migrants travelling long distances and crossing ecological barriers experience different constraints in terms of time, energy and safety than short-/medium-distance migrants without barrier-crossings. As such, natural selection shapes the aerodynamic properties of these groups differently. Yet, to the best of our knowledge, we lack information on whether natural selection has contributed to reducing energetic flight costs through generally lower body mass in trans-Saharan migrants. To fill parts of this gap, we investigated this eco-morphological pattern in 5,410 individuals of 22 Palearctic songbird species ranging from short-/medium-distance to trans-Saharan migrants. We used individual size-independent scaled lean body mass values based on wing length as a measure of body size and, for the first time, precisely determined lean body mass values by direct measurements via quantitative magnetic resonance technology. Scaled lean body mass for a given body size was significantly higher in short-/medium-distance migrants than in trans-Saharan migrants. While scaled lean body mass significantly decreased with increasing migration distance in short-/medium-distance migrants, no such effect was found in trans-Saharan migrants. Our results thus show an eco-morphological pattern relating species’ lean body mass not only to migration distance but also to migration group. This suggests that selective effects of the presence/absence of ecological barriers and/or of a threshold level for migration distance on migrant birds may be more important than the linear continuum of migration distance per se.

During spring migration 2017–2019 and autumn migration 2017–2018, 5,410 individuals of 22 Palearctic songbird species were routinely caught in the trapping garden of the Institute of Avian Research ‘Vogelwarte Helgoland’ on Helgoland (54°11’N, 07°55’E).

Body mass (±0.1 g) using an electronic balance and maximum wing length (i.e. maximum chord, ±0.5 mm), as a measure of body size, were taken for each bird. Hereafter, we used the EchoMRI™ (EchoMRI Body Composition Analyser E26-262-BH, Zinsser Analytic GmbH, Frankfurt am Main, Germany) to directly measure for each bird (1) QMR fat mass, i.e. all fat molecules, and (2) QMR lean mass, i.e. the sum of all water-containing body parts. QMR fat mass was subtracted from the actual body mass to calculate the individual lean body mass. For each species, mean ± standard deviation was determined for actual body mass and lean body mass (g) as well as wing length (mm).

We classified birds according to two main migration groups: (1) short-/medium-distance migrants (pre-Saharan migrants) and (2) long-distance migrants (trans-Saharan migrants). Mean migration distance was estimated between the centre of the breeding and wintering area for each species (rounded up to the nearest 50 km).

One special case is the Oenanthe oenanthe L. (Northern Wheatear) because two subspecies with significantly different migration distances pass Helgoland during migration. We generated the mean migratory distance for each subspecies (leucorhoa = 7,750 km; nominate = 5,850 km) and then used the average distance of these two (= 6,800 km).

Additionally, we were able to sample the local breeding population of the House Sparrow. Ít usually stays on the island year-round and can, therefore, be considered “resident”. As many “resident” and pre-Saharan species show partial migration, where transition between dispersal movements and true migration is flexible, and only one resident species was considered in this study, we included the House Sparrow with an average migration distance of 0 km in the group of pre-Saharan migrants.