During migration, the critical costs and benefits for birds are foremost considered in terms of time, energy and predation risk (e.g. Alerstam and Lindstr6m 1990, Houston 1998). Recently, as in evolutionary ecology in general, also costs of parasitism and immune defence have been added to this list of factors that govern avian migration (Moller and Erritzoe 1998, Figuerola and Green 2000, Waldenstr6m et al. 2002). Although it was regarded as a potential danger to migrants at a relatively early stage (Yapp 1956, 1962), the potential threat of water imbalance during migration has seldom been considered a truly relevant factor that could potentially constrain a large proportion of long-distance migrants. Stimulated by this issue's publication of Sapir et al.'s (2004) study on the effect of free water on fuel-deposition rate in blackcaps Sylvia atricapilla and lesser whitethroats S. curruca in the Negev desert, I would like to briefly review the current status of Yapp's hypothesis that migrants' water budgets may constrain their migratory performance. For a bird, as for many other organisms, inadequate hydration is incompatible with life. Numerous observations exist of birds eagerly ingesting water upon arrival at a stopover site (e.g. Biebach 1990, Pennycuick et al. 1996), a phenomenon that is also well known from pigeon racing. Despite these suggestive observations and the severe consequences of dehydration, the risk of dehydration is not perceived by the majority of ornithologists to be a major and serious constraint during longdistance migration, not even when deserts or oceans have to be crossed. An important impetus for disregarding this potential constraint has been the fact that lowered body-water contents have only rarely been observed with sufficient certainty in long-distance (trans-desert) migrants (Biebach 1990, 1991, Landys et al. 2000). However, various studies on birds and mammals have shown that body water content is in fact a poor indicator of dehydration stress, since it tends to remain stable in animals dying from water restriction (Chew 1951, 1961, Dawson et al. 1979). With this important argument, on which researchers previously have refuted the possibility that a migrant's life is at least partly controlled by the risk of water imbalance (e.g. Biebach 1990, 1991, Landys et al. 2000) becomes invalid. I think that the scope for water stress as an important determinant of migratory behaviour is potentially much larger than most ornithologists are inclined to believe nd I would like to argue that we have been too focussed on finding birds that have been in a physiological state of dehydration, and not on behaviour directed at avoiding dehydration and optimising water balance. Although physiological adaptations in metabolic rate, evaporative water loss and body temperature maintenance may play a substantial role (e.g. Tieleman et al. 2003), problems of water stress in desert birds are generally thought to be solved mainly through behavioural strategies (e.g. Maclean 1996, Williams et al. 1999). If maintaining water balance was the only constraint, the three simple rules to live along in a desert are: stay out of the sun, find yourself a cool place and take it easy during the hottest parts of the day. Migrants may adopt the very same rules of thumb to avoid dehydration. During the hottest parts of the day they might for instance choose to land and find a shaded resting spot, as many birds in fact do while crossing deserts. You may find them sitting immobile in little crevasses, car tracks, or in the protection of anything else that may provide a little bit of shade. Thus, even in the absence of any direct physiological signs of dehydration, the behaviour of migrants may well be governed by the threat of dehydration. Although alternative explanations are possible (e.g. Kerlinger and Moore 1989, Schwilch et al. 2002), the landfall of many birds in the Sahara desert during daytime (Biebach 1990, Biebach et al. 1991, 2000) might be a behavioural strategy to avoid water imbalance. Using the altitudinal profiles of wind, temperature, pressure, and humidity Klaassen and Biebach (2000), and Liechti et al. (2000) tried to explain the altitudinal distributions of nocturnal migrants recorded by radar above deserts in Egypt and Israel. To this end they used