S15.2: Habitat utilisation and energy storage in passerine birds during migratory stopover

Swiss Ornithological Institute, CH-6204 Sempach, Switzerland, fax 41 41 462 97 10, e-mail jennil@orninst.ch

Migrant birds cover enormous distances during their journeys to the wintering grounds and back to the breeding area. Their course of migration seems to be controlled largely by an endogenous spatio-temporal programme (Berthold 1996; initial predictive information sensu Wingfield et al. 1990). Migratory birds, in particular long-distance migrants, show an astonishing spatial and temporal precision in their migration, returning to the same site within a time frame of only a few days each year. This is particularly remarkable since migrants are exposed to a large number of predictable and unpredictable environmental influences, such as weather, changing food availability etc (supplementary, synchronising and modifying information sensu Wingfield et al. 1990). It is well known that migrant birds are able to react to environmental effects, particularly weather and food availability (Bibby et al. 1976; Richardson 1990). It seems, therefore, conceivable that birds possess certain strategies (or rules of reaction) to the current (and possibly future) environmental situation during their migratory journey which complement the endogenous spatio-temporal programme and help to successfully implement it.

However, it is still unclear to what extent the endogenous spatio-temporal programme can be modified. It is even more unclear what the importance and complexity of strategies to environmental situations are. On the one hand, it is conceivable that birds migrate with only a few simple reaction rules to the most important environmental factors (e.g. land when it rains, start when fat stores have attained the programmed level or when there is no food), but otherwise adhere quite strictly to their endogenous spatio-temporal programme. On the other hand, it may be imaginable that the endogenous spatio-temporal programme is only giving a very broad frame and much of the actual migration is developing through the interaction of environmental cues and the physiological state of the bird, mediated by complex reaction rules.

Most migrants need to interrupt their migratory journey at intervals for refuelling in order to reach their goal, particularly before crossing ecological barriers. Hence, their migration is divided into phases of flight and comparatively much longer phases of stopover, the latter determining largely the total duration of migration. During such stopovers a series of ‘decisions’ by the bird (not meant as a conscious choice, see Ens et al. 1994) influence the schedule and the success of its journey. For instance, a bird has to find and select a suitable habitat, to forage in an unknown area, to compete with resident birds and to avoid being predated. Furthermore the bird has to decide how long it shall stay and how much fat it shall deposit. All these decisions might be influenced by a variety of environmental factors. Strategies and ‚decisions‘ taken at stopover sites largely determine the options available during the next flight bout, e.g. maximum flight distance.

Theoretical models are available to explore the effects of certain ‚decisions‘ and strategies and to determine theoretically the optimal behaviour (e.g. Alerstam & Lindström 1990; Weber et al. 1994). These models examine the optimal behaviour with respect to certain specific criteria: either minimising time of migration, energy expenditure or predation risk. The aim of the bird, however, is to survive the period of migration as best as it can and with the best position for the next reproductive period which includes that it stays in the right place at the right time. The resulting spatio-temporal programme is ultimately shaped by selective forces. Such an endogenous programme might include a slow migration or a temporary stay until conditions further along the migration route become favourable (e.g. Marsh Warblers Acrocephalus palustris stopping over for several weeks in NE Africa; Pearson & Lack 1992). The theoretical models include parameters such as stopover duration, rate of body mass inrease and departure fuel load. However, there are many more environmental factors and ‚decisions‘ to be considered. Furthermore, decisions are likely to be dependent on the geographical situation and time of year, and they are interdependent on each other.

Compared with theoretical models, evidence from field studies lags behind. The reason is that field studies with rapidly moving animals are difficult to carry out. Small landbirds are mostly inconspicuous and not readily amenable to direct observation or repeated capture. Despite this, only field studies will reveal the actual rules of decision and their mechanisms.

The aim of this paper is to summarise field data about the behaviour of passerine birds during stopover and refuelling as a reaction to environmental factors, in order to demonstrate the gaps and suggest further studies. We include aspects of habitat selection, landing behaviour, stopover duration, temporary home range, territoriality and refuelling rate of migratory passerines. Passerines migrating over continental Europe can be divided into typical night migrants and typical day migrants, with only a few species migrating regularly during both night and day (Winkler 1984). At the coast and over the sea, however, day migrants may be forced to migrate during the night and night migrants during the day. This paper is concerned with typical night migrants only.

Compared with the breeding and wintering seasons, there are three special features of habitat selection during the migratory journey: (1) the bird is usually staying at a stopover site for only a short period and thus has only limited time for exploration; (2) because of the need to cross less favourable regions between the breeding and wintering site, many migrants at some stage during their journey cannot avoid staying in less favourable or inappropriate habitats for at least a short time; (3) the habitat has to enable survival only (and not reproduction, as during the breeding season), but also the replenishment of energy stores for further migration (as opposed to the wintering season). Apart from these special features, habitat selection during stopover shares many features with habitat selection during the breeding and non-breeding seasons on which a largy body of literature exists which will not be reviewed here. Moreover, habitat selection by nonbreeding migrants has been extensively reviewed by Hutto (1985). Hence, we will focus on the special features of habitat selection during stopover.

It can be assumed that birds are endogenously programmed to fly in a certain direction (Berthold 1996) which exposes them to a given set of habitats along their migration route (Hutto 1985). At the same time, some innate instructions and cues can be assumed that will guide the bird to a suitable habitat (Hildén 1965). Indeed, provided that there is a certain habitat diversity, most small European landbirds stopover in habitats roughly similar to their breeding habitat. For instance, in central Europe woodland species are largely confined to woods and bushes during stopover and reed-bed species to reeds (Bairlein 1981; Degen & Jenni 1990; Streif 1991; Mädlow 1994; Jenni & Widmer 1996). However, even though the breeding habitat is available, some species use different habitats during migration. This is especially true for species preferring woods or open woodland, while reed-bed species are more strictly confined to reeds. Particularly species of bushy or open wooded breeding habitats may invade reed-beds, and species typical of interior woodland during the breeding season use wood edges and bushy areas during the autumn migration season (Bairlein 1981, 1983; Jenni & Widmer 1996). At a given stopover site, migrants consistently use a certain habitat type. This results in an ecological separation of most of the species migrating concurrently (Bairlein 1981, 1983).

The situation is different when the breeding habitat is not available in the area of stopover. Then, migrants are forced to show great flexibility in habitat preferences (cf. Leisler 1992). It is unknown whether birds which use a different habitat during migration than the breeding habitat or whose breeding habitat does not exist along the migratory route have two different innate instructions on what habitat to look for or whether their innate instructions incorporate both habitat types.

There are a number of conceivable factors which might guide a migrant to land in a certain area and to stay in that area or to leave it, e.g. habitat structure, food availability, predation risk, intra- and interspecific competition, weather and microclimate. There are examples in the literature that birds do react to each of these factors (e.g. Bibby et al. 1976; Hutto 1985; Bibby & Green 1980, 1983; Greenberg 1986; Biebach 1990; Lindström 1990; Fransson & Weber 1997).

However, there is virtually no study investigating whether - and if so how - migrants at stopover sites acquire knowledge of their surroundings and assess and compare the quality of two or more areas. Effectively, a migrant landing in a certain area has a difficult choice: it might invest time for assessing habitat quality within a certain perimeter around its landing point, to fine tune its habitat selection, and perhaps to acquire a temporary territory. This depends not only on a trade-off between costs involved with searching for a better habitat and benefits such as a higher refuelling rate or reduced predation risk at the current stopover site, but also on the probable situation at a stopover site forward or backward along the migration route. Hence in the case of migrating birds, the decision to stay or to leave a current site can be considered as part of the assessment of habitat quality. For instance, in the case that possible gains are minimal (e.g. uniform habitat quality over a large area), assessing habitat quality is a waste of time and should be avoided. If habitat quality at the current site and its surroundings is poor, but likely to be better at a distant site, assessing habitat quality at the current site is also not fruitful and the bird should leave. There may also be situations in which assessing habitat quality is too costly, too time consuming or too risky and birds may either accept an incomplete knowledge of the quality of their area and stay where they land or leave the area and migrate further. On the other hand, assessing and comparing habitat quality might be highly profitable and an ongoing process over the entire stopover period.

In any case, a migrant bird stopping over finds itself in an unfamiliar environment. However, a bird benefits by familiarising itself with an area through e.g. more efficient foraging and escaping predators more efficiently (e.g. Stephens & Krebs 1986). Therefore, at a smaller scale than assessing habitat quality, we would expect migrants to familiarise themselves with the immediate area where they stopover.

There are basically two types of information and two means to acquire knowledge about the surroundings: primary information through exploration and secondary information through reaction to indirect cues of habitat quality.

Exploration after landing involves active displacements and thus the birds can get primary information about food availability, shelter from predators and other resources. Exploration is time-consuming and may involve a high predation risk. For small migrant landbirds during stopover, there is little evidence for exploration behaviour, but such evidence is very difficult to produce. Some of the Summer Tanagers Piranga rubra followed after arrival at an island with radio-telemetry showed a movement pattern consistent with exploration, but others not (Aborn & Moore 1997).

Instead of engaging in a lengthy and risky exploration at the stopover site, birds might use secondary cues which are quickly perceived. Possible secondary cues are habitat structure, the presence of conspecifics and the presence of syntopic animals.

The fact that the great majority of birds is found in their typical habitats early in the morning indicates that night migrants generally land in an appropriate habitat. Because there is no movement during the night, the landing spot must be selected by visual cues of habitat structure (see below). Tape-luring of migrants also indicates that the song of conspecifics and of other bird species inhabiting the same habitat are a cue for landing night migrants (see below). Although some species are territorial during stopover, others are often caught in loose flocks (unpubl. obs.). Hence, after landing, an attraction of migrants to conspecifics or heteropecifics during the day might not be excluded.

Migrants may also acquire some knowledge of the landscape while flying during the night. This may be viewed as perceiving habitat quality over large distances via secondary cues without extra costs. There is good evidence that lean birds engage in reverse migration when they encounter the sea (Åkesson et al. 1996). This might be the best strategy when migrants need to refuel before crossing an ecological barrier whose extension they cannot foresee or when they cannot forecast the quality of stopover sites beyond. Also birds which set off to cross the sea may return to an island or the cost which they have visited or overflown shortly before. At the Balearic Island Mallorca and at the south coast of Spain, 8% and 3%, respectively, of the birds aloft 2 h after the start of nocturnal migration were reverse migrants which landed again. These figures increased to 20% and 15%, respectively, towards the end of the night (Bruderer & Liechti in press). In summary, there is substantial evidence that birds use secondary cues on a smaller and larger scale for selecting an appropriate stopover site.

Migrants during stopover may often find themselves in suboptimal habitats, either because no optimal habitat is available or because their ability to find the optimal habitat is limited. This may be a reason why migrant species are more flexible in their foraging behaviour and more uniform in their morphology than resident species (reviewed in Leisler 1990).

In order to better understand the processes of habitat selection during stopover, some basic information is needed on landing behaviour, stopover duration, temporary home range and territoriality, which will be dealt with in the following sections. Subsequently, the evidence for possible costs of an exploration behaviour, i.e. a reduction in refuelling rate after landing, is discussed, and finally some aspects of the metabolism of refuelling.

Night migrating passerines usually start flight 0.5 - 2.5 h after sunset (Sutter 1957; Able 1970; Bruderer 1971; Gauthreaux 1971; Alerstam 1976; Bruderer & Liechti 1995; Bruderer et al. 1996; Liechti et al. 1997). However, there are indications, that some birds may start later during the course of the night (Cochran et al. 1967; Hebrard 1971; Åkesson 1995). A large proportion of birds migrating over continental Europe and North Africa end a nightly flight bout during the night and land (e.g. Biebach et al. 1991). This is confirmed by a strong decrease of the number of night migrants observed by radar in Switzerland, Germany and Israel, usually from midnight onwards (Sutter 1957; Bruderer 1971; Bruderer & Liechti 1995; Bruderer et al. 1995). In the Negev, the number of birds aloft drops earlier at night in autumn than in spring, possibly because the habitat becomes dryer and dryer along the migratory route in autumn, but better and better in spring (Bruderer & Liechti 1995). Over land, all night migrants interrupt their nocturnal journey at sunrise at the latest. Therefore, a large proportion of night migrants land during the dark hours.

During autumn migration in Europe, almost all nocturnal migrants are observed in their typical habitats, provided that they are available in that region (e.g. Bairlein 1981, 1983; Degen & Jenni 1990; Jenni & Widmer 1996). Therefore, nocturnal migrants must be able to recognise their habitat at night. Optical and acoustic cues enable the birds to find their appropriate habitat. Herremans (1990) reported, that birds were attracted by playing tapes with familiar vocalisations of woodland or marshland species, respectively, but not by unknown species. The choice pattern in favour of the ‘correctly inferred habitat’ is significant, thus corroborating the ability of night migrants to assess habitat by conspecific and interspecific recognition of vocalisations. The preference of Marsh Warblers for a marsh site near a woodlot and of the Reed Warblers Acrocephalus scirpaceus for a marsh site away from the wood implies also a visual assessment of habitat structure at night (Herremans 1990). The importance of visual cues is confirmed by investigations of habitat selection when visibility is impaired due to ground fog (Jenni 1996). Under these circumstances, increased numbers of migrant species depending on wood and bushes were found in the reed-bed which they first met after crossing a lake.

Birds landing during the night apparently stay at the landing spot until first light, because there are no captures during night over many years of autumnal catching activities (own obs.). A slightly higher proportion of birds in atypical habitats was observed in the first hour after dawn, indicating that birds specify their habitat selection in the early morning (Bairlein 1981). In a small nature reserve with willow bushes, sedge and reed marshes, Reed Warblers were found in adjacent maize fields only in the early morning, suggesting that they leave them quite promptly after first light (Degen & Jenni 1990). European Robins Erithacus rubecula which landed in the atypical habitat of a reed-bed during nights with ground fog left the reed-bed as soon as the fog dissolved (Jenni 1996).

According to present knowledge, the landing behaviour of small landbirds may be summarised as follows. Night migrants are programmed to land before sunrise, except if there is no appropriate habitat at all (e.g. when flying over water or bare desert; see Biebach et al. 1986 for a detailed hypothesis). Environmental and internal (physiological) factors modify the time and determine the exact place of landing. Birds react to the availability of habitats and select appropriate habitats during the night on visual and acoustic cues. How finely they are able to discriminate habitats during the night from aloft is unknown, but there seems to be a certain degree of readjustment in the very early morning. Inducing landfall by tape-luring during the night (Herremans 1990; Schaub et al. in prep) demonstrates that environmental cues (indicating a favourable habitat) influence migrants to land (even in inappropriate habitats). But there is no information for how long these birds would have migrated further. Apart from information on available habitats, many more factors are certainly involved in the decision to land during the night: night migrants land when energy stores are nearing exhaustion; they land if conditions to continue flight are bad (headwinds, storms and rain (e.g. Moore & Kerlinger 1987), high ambient temperature over desert (Biebach 1990)). In these cases, migrants may land in atypical habitats, either because their typical habitat is not available in the surroundings or because visibility is poor and acoustic cues missing. The kind of factors and the interplay of environmental and internal (physiological) factors deciding about landing as well as the mechanisms involved are only beginning to emerge and deserve further study.

Direct measurements of the time spent at a stopover place are only possible, if individual birds are followed from arrival until departure (e.g. with satellite tracking). In the case of small passerines which are too small for current satellite tracking devices, the time of departure might be observed directly (e.g. in radio-tracked birds), but the arrival time will be known only in exceptional cases. Also the effect of a transmitter on the bird’s stopover duration will be difficult to assess. Therefore, in most cases, stopover duration of small birds will have to be estimated from ringed birds, i.e. from a capture-mark-recapture data-set.

A first approach still often used to estimate stopover duration from capture-mark-recapture data sets is to calculate the time elapsed between first and last capture (minimum stopover duration; e.g. Cherry 1982; Ellegren 1991; Kuenzi et al. 1991; Morris et al. 1994, 1996; Kaiser 1996; Woodrey & Moore 1997). Firstly, this measure admittedly neglects the time a bird may have spent at the stopover site before first capture and after last capture. Second, this measure is based only on those birds caught at least twice and may not be representative for the entire population. Some authors (e.g. Morris et al. 1994; Kaiser 1995) justified this second point by assuming that the birds captured only once were transients. But as the capture probabilities are often very low, the division into transients and residents according to the capture history has a low power and the justification for distinguishing between transients and residents is weak. Often, minimum stopover duration is used as a relative measure for the comparison of two or more groups of birds. However, because the time spent before first capture and after last capture as well as the capture probability might vary among the groups to compare, this is not generally a valid approach.

A second approach is to analyse the capture-mark-recapture data with Jolly-Seber models. This method allows to estimate survival and capture probabilities separately. Because birds usually stay only a few days at a stopover place, it can be assumed that the real survival probability is near one and the estimated survival probability therefore the probability of staying at the stopover place. This probability can be converted into an expected mean stopover duration. Lavee et al. (1991) and Holmgren et al. (1993) found simple formulas for stopover duration when survival and capture probabilities are assumed to be constant over the study period. Kaiser (1995) calculated stopover duration when survival and capture probabilities vary over time. However, the estimates of survival probabilities are always conditioned on first capture. This means that the expected time the birds spend at a stopover place after it has been captured for the first time is estimated. The time the bird spent at the stopover place before first capture remains unknown.

With the advent of refined methods to analyse capture-mark-recapture datasets, it is possible to estimate the probability of being in the population before the current capture, the seniority probability, by means of a recruitment analysis from capture-mark-recapture data (Pradel 1996). This analysis is basically the same as an ordinary survival analysis and offers the same possibilities of hypothesis testing. Analogous to survival probabilities, seniority probabilities can be converted into an estimate of the time the bird spent at the stopover place before first capture. The total, integral stopover duration is therefore the sum of the time the birds spend at the place before first capture and the time the bird spends thereafter (R. Pradel & M. Schaub unpubl.). In non-moulting first-year Reed Warblers during autumn migration at a stopover site in Switzerland, it turns out that the estimated mean integral stopover duration according to this analysis (13.8 days) is 1.8 times longer than mean minimum stopover duration (7.7 days; median 4 days) and 2.0 times longer than the estimated stopover duration after first capture (approach two; 6.9 days) (R. Pradel & M. Schaub unpubl. data). In summary, it is very probable that stopover duration of many passerines migrating over areas offering many stopover habitats is considerably longer than previously thought. This may change our thinking of how important an exploratory phase after landing may be.

Because of the shortcomings of the methods applied until recently, it is difficult to judge whether observations of differences in minimum stopover durations between age classes or sexes (e.g. Ellegren 1991; Lavee et al. 1991; Morris et al. 1996) are real or in fact reflect differences in catching probabilities. Lean birds often seem to be more likely to stopover (higher recapture rate) and to have a longer minimum stopover duration than fat birds, particularly in suboptimal habitats such as desert oases and islands reached after a sea-crossing (e.g. Bairlein 1985a; Biebach et al. 1986; Moore & Kerlinger 1987; Winker et al. 1992; Morris 1996; Morris et al. 1996; Woodrey & Moore 1997; Yong & Moore 1997). Again, further studies have to show whether this is due to a higher capture probability of lean birds (as observed by Bairlein 1985a in oases) or due to a genuine difference in stopover duration between lean and fat birds (as suggested by Safriel & Lavee (1988) for some species at a small oasis where arrival and departure could be fairly well monitored). In Reed Warblers at a Swiss stopover site, integral stopover duration of lean birds was in fact longer than in fat birds (M. Schaub unpubl. data). However, also in lean birds, stopover depends on the suitability of the habitat. Lean birds landing in unsuitable habitats will continue migration (e.g. Biebach et al. 1986; Winker et al. 1992; Woodrey & Moore 1997).

Because of a lack of appropriate estimates of stopover duration, we are only beginning to understand how stopover duration is determined and how stopover duration relates to exploration of the surroundings. First, it is unclear whether physiological factors, such as fatigue or the need for sleep, require a stopover of a certain length. Second, there might be an innate rhythm of flight and stopover periods or of refuelling periods which give the frame of stopover duration. Rhythmic body mass changes of about 2 weeks in Garden Warblers kept under standard conditions may hint on such a phenomenon (Bairlein 1986). Third, many environmental factors may influence the duration of stopover at a particular site, such as initial energy reserves, refuelling rate, further stopover possibilities, time pressure, weather, predation risk, distance to the point of initiation and end of migration (e.g. Bairlein 1985a; Biebach et al. 1986; Lavee et al. 1991; Winker et al. 1992; Morris 1996; Fransson & Weber 1997; Woodrey & Moore 1997).

The size of a short-term, temporary home range during stopover is a trade-off between on the one hand accessing additional food resources and exploring the surroundings and on the other hand increased familiarity with the stopover area and possibly territory defence. Hence, it may be expected that the size of the temporary home range and the social system (e.g. territoriality, non-territoriality or flocking) varies between species and with environmental conditions, for example dependent on the kind of food taken, food availability, the spatial distribution of food, inter- and intraspecific competition, characteristics of the habitat etc. Indeed, some species or individuals have been reported to be territorial at certain staging areas (e.g. Rappole & Warner 1976; Bibby & Green 1980; Greenberg 1986) or defend food resources (e.g. Young 1990), others are not territorial but stay within a temporary home range and others are reported to roam around widely, either singly or in flocks. Northern Waterthrushes Seiurus noveboracensis defend temporary territories in coastal Texas and only territory holders gain body mass, the others depart shortly after arrival regardless of their energetic condition (Rappole & Warner 1976). In Bluethroats Luscinia svecica, dominance (correlated with size) positively influenced refuelling rates (Lindström et al. 1990).

The few studies investigating the size of the temporary home range in non-flocking, small migrants stopping over in favourable habitats found quite small surfaces. One to four Northern Waterthrushes occupied territories for successful refuelling around a pond 150 x 40 m (Rappole & Warner 1976). Temporary home range size of three territorial Pied Flycatchers Ficedula hypoleuca refuelling in Portugal was 1466 – 2906 m² (Bibby & Green 1980). Summer Tanagers arriving from a trans-Gulf crossing moved generally over less than 500 m linear distance during their first days (Aborn & Moore 1997). Garden Warblers Sylvia borin and Blackcaps S. atricapilla at a stopover site in Switzerland were not territorial. The size of their temporary home range, as determined by 18 radio-tracked birds, comprised a mean area of 30000 m² (7000 - 75'000 m²). There was only a slight expansion of the area visited by an individual with increasing stopover duration and no correlation between the area visited and the number of days the birds have been tracked. Essentially the birds remained in the same area, although the habitat extended over a much wider surface and although there was no direct evidence of intraspecific competition, limiting the size of the temporary home range of an individual (L. Jenni & F. Widmer unpubl. data). The few available studies hint to the conclusion, that in many migrants, familiarity with the stopover site, and in some cases dominance or territoriality, is an essential component of successful refuelling. However, more studies with individually marked and followed individuals are needed.

If exploration, becoming familiar with the stopover area and possibly establishing a territory is inflicting a cost, this is likely to translate into a lower refuelling rate at the beginning of the stopover period. While it would be difficult to recognise a positive, but slightly lower than normal fattening rate in practice, an initial decrease in body mass or an initially stable body mass could be taken as an indication of such costs. Consequently, the term initial body mass loss or initial mass loss at arrival is used in the literature (e.g. Lindström 1995). However, because these costs may also translate into a lower, but still positive refuelling rate, the term initial reduction in refuelling rate is used in this paper. In the case birds can stop anywhere along the migration route, the concept of an initial cost at arrival is also basic to the model of optimal bird migration derived from flight mechanical theory, because only the assumption of a search and settling time allows to determine an optimal stopover time and flight range of birds maximising migration speed (Alerstam & Lindström 1990; Weber et al. 1994). An initial cost at arrival theoretically also relates to optimal fat load at departure, frequency of stopovers and overflying suitable stopover sites. However, an initial decrease in refuelling rate can have other causes than costs of getting established at the stopover site.

The main body of the alleged evidence for an initial decrease in refuelling rate comes from birds trapped at least twice at stopover sites which often have been found to lose mass during the first days after being caught (e.g. Winker et al. 1992; Lindström 1995; Yong & Moore 1997). However, the causes of the observed initial loss in body mass are only speculative. They might include effects of trapping and handling and various natural causes of which costs of getting established at the stopover site is only one.

There are three problems with body mass data of recaptured birds. First, in most of these datasets, it is assumed that the birds had arrived the preceding night. However, in most studies this is an unproven assumption. Because stopover duration may be rather long (see above), the proportion of newly arrived birds in a trapping total may actually be often quite low. Second, it is debatable how representative the minority of birds recaptured is for the entire population of migrants. For instance, less successful individuals might roam around more and be more likely recaptured than individuals successfully replenishing in a small temporary home range. Third, an initial decrease in refuelling rate may be due to effects of trapping and handling. These include handling shock and lost opportunities to feed by keeping birds in captivity and by releasing birds at a place different from their catching place. It is obvious that catching and keeping birds precludes them from feeding and the question is how important the lost feeding opportunities are and how quickly the birds can compensate for.

Besides studies showing an initial decrease in body mass, there are others with conventional trapping and handling procedures demonstrating that most migrants gained mass from the first day after capture (e.g. Cherry 1982; Carpenter et al. 1993; Bairlein 1987; Moore & Kerlinger 1987). Other studies reported a decrease in body mass after capture in young birds only, but not in adults (e.g. Ellegren 1991). In some studies, there was no correlation between how often a bird was caught and its body mass development (e.g. Yong & Moore 1997). Such studies are cited as support for the general assumption that trapping has no effect on refuelling. However, this might not be true. First, trapping may induce a lower, but still positive change in body mass. Second, an effect of trapping might depend on the particular condition of the bird and on the conditions of the study site. For instance, high food availability or dominance may allow certain birds to compensate more quickly (e.g. within a day and, hence, unnoticeable to the observer) than others. Similarly, studies showing a decrease in body mass after capture in sedentary birds (e.g. Åkesson et al. 1995) may not be readily transferable to migrants.

Recently, we analysed data of European Robins for changes in body mass after they have been caught twice; thereby we were sure that the birds had not arrived the night before at the stopover site in Switzerland. They showed a decrease in body mass during the first 1-2 days after retrapping, suggesting an effect of trapping (R. Schwilch unpubl. data). In order to study body mass changes after arrival, we tape-lured Reed Warblers into a small reed-bed and estimated the refuelling rate from plasma metabolites (cf. Jenni-Eiermann & Jenni 1994) collected within a few minutes after trapping, i.e. before a possible trapping effect. There was no noticeable initial decrease in refuelling rate (R. Schwilch unpubl. data). In contrast, Rappole & Warner (1976) observed slightly decreasing body mass in Northern Waterthrushes which did not have a territory and increasing body mass in territory holders, suggesting that birds which have to wait for a territory incur an initial cost at arrival.

In summary, the evidence for an initial reduction in refuelling rate after arrival in small passerines is still meagre. With the exception of that of Rappole & Warner (1976), we know of no published study relating body mass development to the time of arrival of individual birds, studies which are badly needed. It may well be that the occurrence of an initial reduction in refuelling rate after arrival depends on the conditions at the stopover site. For instance it may be more noticeable if feeding conditions are more stringent than in good feeding conditions.

If an initial reduction in refuelling rate after arrival actually occurs, it may have several causes: competition among conspecifics (including the establishment of a territory), inefficient foraging because of unfamiliarity with the stopover habitat (including exploration) and physiological reasons. Although the two former are obvious reasons and there is circumstantial evidence, there is a lack of conclusive studies, which exclude possible trapping effects.

There is now substantial evidence for physiological reasons from laboratory experiments and field observations. Rufous Hummingbirds Selasphorus rufu during their migration from the Pacific Northwest to their wintering sites in Mexico showed a biphasic mass gain, a slow one up to about 3.5 g and a steeper one thereafter (Carpenter et al. 1993). The authors hypothesised that during the slow first phase muscles are rebuilt and that the higher energy costs of synthesising protein than fat might lead to a slower rate of mass gain. After a period of starvation in the laboratory, Garden Warblers and Thrush Nightingales Luscinia luscinia were increasing in body mass more slowly during the first day of refeeding than during subsequent days (Klaassen & Biebach 1994; Klaassen et al. 1997). This suggests that a reduction in digestive organs (intestine, stomach, liver), as was indeed observed (Hume & Biebach 1996), prevented the birds from processing large amounts of food. In shorebirds covering very long distances non-stop, a reduction in digestive organs was found even prior to departure (Piersma & Gill 1998).

Whether it is muscle or digestive organs reduced before or during a migratory flight, the reduction of tissue protein is likely to be proportional to flight distance. Therefore, it can be regarded as part of the restorage of body energy stores after a flight. This is in contrast to initial costs at arrival involved with exploration and settling which can be regarded as discrete costs occurring at each stopover independent of the length of the previous flight. The hypotheses put forward to explain naturally occurring initial reductions in refuelling rate do not rule out each other. It is likely that they may be of different importance between species and individuals and between stopover sites.

Whether or not a migrant suffers an initial reduction in refuelling rate, the refuelling rate is an important parameter in migration ecology. Especially in night migrants, it largely determines the time spent on migration and, according to optimal migration theory (Alerstam & Lindström 1990), it is expected to determine the time spent on a particular stopover site in time-selected migrants. It certainly depends on food availability, but also on other environmental factors such as competition, dominance, predation risk and weather. For instance, birds may avoid areas of high predation risk despite a very high food availability or they might reduce the refuelling rate in order to lower predation risk (Lindström 1990; Fransson & Weber 1997). Subordinate birds may suffer a reduced refuelling rate (e.g. Lindström et al. 1990). In cold or windy weather, foraging success and consequently refuelling rate may be lower (own unpubl. data). In optimal feeding conditions, there is a physiological upper limit to the amount of metabolisable energy intake with consequences on mass-dependent refuelling rates (see Kirkwood 1983; Lindström 1991, 1995).

How do birds manage to build up energy reserves for migration? Hyperphagia is regarded as the principal mechanism (e.g. Berthold 1975), which results in hyperlipogenesis. The surplus energy intake is transformed and deposited as fat in adipose tissue, independently of whether the surplus is derived from carbohydrates, protein or fat.

A factor influencing refuelling rate is moult. There are many studies showing that high refuelling rates do not occur in moulting birds (e.g. Winker et al. 1992; Jenni & Jenni-Eiermann 1996). It is yet unclear, whether this involves a physiological or metabolic incompatability (e.g. between protein metabolism for moult and fat metabolism for energy storage) or whether energy demands for moult consume the surplus energy which could be deposited.

Hyperphagia may be associated with physiological and metabolic adaptations. For instance, an increase in the mass of the digestive organs, together with an increase in basal metabolic rate, may be expected (cf. Piersma & Lindström 1997), but has not been shown so far in passerines. An increased efficiency of food utilisation, particularly in the digestibility of fat and carbohydrate was observed in an experimental study with hyperphagial Garden Warblers (Bairlein 1985b, but see Klaassen & Biebach 1994). Another proposed factor contributing to fat depostion is reduced activity (Blem 1980). In an experimental study with Garden Warblers (Klaassen & Biebach 1994) a reduced activity was measured during the recovery period after a fasting period of 2-6 days, and it was calculated, that these energy savings can explain about 30% of the body mass increase during recovery.

Metabolic studies in free-living passerines gave insight into their fat metbolism. During the autumn migration period at a stopover site in Switzerland, plasma triglyceride levels of Blackcaps, Garden Warblers and Robins were generally higher than during the postbreeding and moulting period and also showed a steeper increase of triglyceride levels during daylight hours (Jenni & Jenni-Eiermann 1996; Jenni-Eiermann & Jenni 1996). Increased lipid storage during the day is correlated with high plasma triglyceride levels (Jenni-Eiermann & Jenni 1994). Therefore, the high triglyceride levels observed in feeding birds as well as their higher body masses during the migratory period reflect increased lipid storage during the day associated with migratory fattening. Migratory birds also seem to spare their fat deposits during nocturnal fasting. Normally, lipids from adipose tissue are the major energy source during fasting (Cherel et al. 1988) and, hence, increased free fatty acid and glycerol levels, both indicators of fat catabolism (Jenni-Eiermann & Jenni 1991), were to be expected in overnight fasted birds. However, in contrast to the postbreeding birds, migratory birds did not rely on lipids from adipose tissues, as indicated by the significantly decreased triglyceride levels. It seems, that they rely to a large extent on plasma triglycerides. A major source of plasma triglycerides during overnight fasting is probably the liver, the main site of lipid synthesis during times of fat storage. Additionally, triglycerides stored in skeletal muscles may form an endogenous energy source. Indeed, the lipid content of the liver and flight muscles and its overnight drop generally increase from the postbreeding and moulting to the migratory period in small passerines (Farner et al. 1961; Dolnik & Blyumenthal 1967; Bagott 1977; Pilo & George 1983). Hence, migratory birds have, due to the increased fat deposition rate, a concomitant hepatic and muscular evening lipid store, which allow a higher utilisation of plasma triglycerides and a lower utilisation of triglycerides derived from adipose tissue.

Not surprisingly, this overview demonstrates a lack of field studies in many fields of stopover ecology of small landbirds. Field studies will generate many more interesting observations and, hence, hypotheses to be incorporated in the current theoretical models of optimal migration.

The concept to distinguish between processes determined by the endogenous programme (or initial predictive information, sensu Wingfield et al. 1990) and processes determined by modifying environmental factors (supplementary, synchronising and modifying information sensu Wingfield et al. 1990) will in our opinion be helpful in the design of future field studies. It may help to disentangle the actual mechanisms taking place and get over the stage of correlative studies.

At the more practical level, it is important to realise that trapping, a method widely used in the study of small inconspicuous landbirds living in dense vegetation, may create artifacts which are difficult to separate from the phenomena to be observed. This particularly concerns the development of body mass in recaptures. Another difficulty in conventional trapping studies is that the time of arrival (i.e. the history of a trapped bird) is generally unknown. Hence, many of the special features of a migrant stopping over for a short time cannot be related to its arrival.

It may also be useful to recall the variability of the environment along the migration route and from year to year. The decisions taken by birds may vary dramatically between the spring and autumn migration seasons, along the migratory route and with progressing migration season. For instance, migrants may place different priorities in the autumn than in spring (e.g. Lavee et al. 1991; Winker et al. 1992). Migrants may overfly or leave stopover sites more readily close to their destination than far from their destination (e.g. Lavee et al. 1991). Also the geographical situation of a stopover site relative to an ecological barrier is an important factor to be kept in mind (e.g. Bairlein 1985a; Safriel & Lavee 1988; Lavee et al. 1991; Åkesson et al. 1996). However, the conditions at stopover sites may also vary between sites at short range. This is especially important to consider when findings obtained at one site are taken to be representative for an entire region. As in any other field study, results may differ between years, evoked by variation in weather or food availability (e.g. Bibby et al. 1976; Moore & Kerlinger 1987; Winker 1995). We feel that the variability of the environment along the migration route and between years offers the potential for many more comparative studies.

We benefitted from a sabbatical stay at the Department of Zoology at the University of Washington in Seattle. We thank Michael Schaub and Regine Schwilch for unpublished results and critical comments on the manuscript.

Aborn, D.A. & Moore, F.R. 1997. Pattern of movement by Summer Tanagers (Piranga rubra) during migratory stopover: a telemetry study. Behaviour 134: 1077-1100.

Able, K.P. 1970. A radar study of the altitude of nocturnal passerine migration. Bird Banding 41: 282-290.

Åkesson, S, Karlsson, L., Walinder, G. & Alerstam T. 1996. Bimodal orientation and the occurrence of temporary reverse bird migration during autumn in south Scandinavia. Behavioural Ecology and Sociobiology 38: 293-302.

Åkesson, S. 1995. Avian migratory orientation: geographic, temporal and geomagnetic effects. PhD Thesis, University of Lund, Lund, Sweden.

Alerstam, T. 1976. Nocturnal migration of thrushes (Turdus spp.) in southern Sweden. Oikos 27: 457-475.

Alerstam, T. & Lindström, Å. 1990. Optimal bird migration: the relative importance of time, energy and safety. In: Gwinner, E. (ed) Bird migration; Berlin; Springer: 331-351.

Bagott, G.K. 1977. Changes in liver and blood composition of the premigratory Willow Warbler in autumn. Comparative Biochemistry and Physiology 56A: 461-466.

Bairlein, F. 1981. Ökosystemanalyse der Rastplätze von Zugvögeln: Beschreibung und Deutung der Verteilungsmuster von ziehenden Kleinvögeln in verschiedenen Biotopen der Stationen des „Mettnau-Reit-Illmitz-Programmes’. Ökologie der Vögel 3: 7-137.

Bairlein, F. 1983. Habitat selection and associations of species in European passerine birds during southward, post-breeding migrations. Ornis Scandinavica 14:239-245.

Bairlein, F. 1985a. Body weights and fat deposition of Palaearctic passerine migrants in the central Sahara. Oecologia 66: 141-146.

Bairlein, F. 1985b. Efficiency of food utilisation during fat deposition in the long-distance migratory garden warbler, Sylvia borin. Oecologia 68:118-125.

Bairlein, F. 1986. Spontaneous, approximately semimonthly rhythmic variations of body weight in the migratory garden warbler (Sylvia borin Boddaert). Journal of Comparative Physiology B 156: 859-865.

Bairlein, F. 1987. The migratory strategy of the Garden Warbler: a survey of field and laboratory data. Ringing and Migration 8: 59-72.

Berthold, P. 1975. Migration: control aqnd metabolic physiology. In: Farner, D.S., King, J.R. & Parkes, K.C. (eds) Avian Biology. Vol. 5; New York; Academic Press: 77-128.

Berthold, P. 1996. Control of bird migration. London; Chapman&Hall: 355pp.

Bibby, C.J. & Green, R.E. 1980. Foraging behaviour ofd migrant pied flycatchers, Ficedula hypoleuca, on temporary territories. Journal of Animal Ecology 49: 507-521.

Bibby, C.J. & Green, R.E. 1983. Food and fattening of migrating Warblers in some French marshlands. Ringing and Migration 4: 175-184.

Bibby,C.J., Green, R.E., Pepler, G.R.M. & Pepler, P.A. 1976. Sedge Warblers migration and reed aphids. British Birds 69: 384-398.

Biebach, H. 1990. Strategies of trans-Sahara migrants. In: Gwinner, E. (ed) Bird migration; Berlin; Springer: 352-367.

Biebach, H., Friedrich, W. & Heine, G. 1986. Interaction of body mass, fat, foraging and stopover period in trans-saharan migrating passerine birds. Oecologia 69: 370-379.

Biebach, H., Friedrich, W., Heine, G., Jenni, L., Jenni-Eiermann, S. & Schmidl, D. 1991. The daily pattern of autumn bird migration in the northern Sahara. Ibis 133: 414-422.

Blem, C.R. 1980. The energetics of migration. In: Gauthreaux Jr, S.A. (ed) Animal migration, orientation and navigation; New York; Academic Press: 175-224.

Bruderer, B. 1971. Radarbeobachtungen über den Frühlingszug im schweizerischen Mittelland. Der Ornithologische Beobachter 68: 89-158.

Bruderer, B. & Liechti, F. 1995. Variation in density and height distribution of nocturnal migration in southern Israel. Israel Journal of Zoology 41: 477-487.

Bruderer, B., Underhill, L.G. & Liechti, F. 1995. Altitude choice by night migrants in a desert area predicted by meteorological factors. Ibis 137: 44-55.

Bruderer, B. & Liechti, F. (in press) Flight behaviour of nocturnally migrating birds in coastal areas - crossing or coasting. Journal of Avian Biology.

Bruderer, B., Liechti, F. & Steuri, T. 1996. Migrations aviennes à travers l’ouest méditerranéen – direction de vol au printemps au-dessus de Majorque. Alauda 64: 7-16.

Carpenter, F.L., Hixon, M.A., Beuchat, C.A., Russell, R.W. & Paton, D.C. 1993. Biphasic mass gain in migrant hummingbirds: body composition changes, torpor, and ecological significance. Ecology 74:1173-1182.

Cherel, Y., Robin, J.-P. & Le Maho, Y. 1988. Physiology and biochemistry of long-term fasting in birds. Canadian Journal of Zoology 66: 159-166.

Cherry, J.D. 1982. Fat deposition and length of stopover of migrant White-crowned Sparrows. Auk 99: 725-732.

Cochran, W.W., Montgomery, G.G. & Graber, R.R. 1967. Migratory flights of Hylocichla thrushes in spring: a radiotelemetry study. Living Bird 6: 213-225.

Degen, T. & Jenni, L. 1990. Biotopnutzung von Kleinvögeln in einem Naturschutzgebiet und im umliegenden Kulturland während der Herbstzugzeit. Der Ornithologische Beobachter 87: 295-325.

Dolnik, V.R. & Blyumenthal, T.I. 1967. Autumnal premigratory and migratory periods in the Chaffinch (Fringilla coelebs coelebs) and some other temperate-zone passerine birds. Condor 69: 435-468.

Ellegren, H. 1991. Stopover ecology of autumn migrating Bluethroats Luscinia s. svecica in relation to age and sex. Ornis Scandinavica 22: 340-348.

Ens, B.J., Piersma, T. & Tinbergen, J.M. 1994. Towards predictive models of bird migration schedules: Theoretical and empirical bottlenecks. NIOZ-Rapport 5, Texel, The Netherlands.

Farner, D.S., Oksche, A., Kamemoto, F.I., King, J.R. & Cheyney, H.E. 1961. A comparison of the effect of long daily photoperiods on the pattern of energy storage in migratory and non-migratory finches. Comparative Biochemistry and Physiology 2: 125-142.

Fransson, T. & Weber, T.P. 1997. Migratory fuelling in blackcaps (Sylvia atricapilla) under perceived risk of predation. Behavioural Ecology and Sociobiology 41: 75-80.

Gauthreaux Jr, S.A. 1971. A radar and direct visual study of passerine spring migration in southern Louisiana. Auk 88: 343-365.

Greenberg, R. 1986. Competition in migrant birds in the nonbreeding season. Current Ornithology 3: 281-307.

Hebrard, J.J. 1971. The nightly initiation of passerine migration in spring: a direct visual study. Ibis 113: 8-18.

Heeremans, M. 1990. Can night migrants use interspecific song recognition to assess habitat? Le Gerfaut 80:141-148.

Hildén, O. 1965. Habitat selection in birds. Annales Zoologici Fennici 2: 53-75.

Holmgren, N., Ellegren, H. & Pettersson, J. 1993. Stopover length, body mass and fuel deposition rate in autumn migrating adult Dunlins Calidris alpina: evaluating the effects of moulting status and age. Ardea 81: 9-19.

Hume, I.D. & Biebach, H. 1996. Digestive tract function in the long-distance migratory garden warbler, Sylvia borin. Journal of Comparative Physiology B 166: 388-395.

Hutto, R.L. 1985. Habitat selection by nonbreeding, migratory land birds. In: Cody, M.L. (ed) Habitat selection in birds; London; Academic Press: 455-476.

Jenni, L. 1996. Habitatwahl nachtziehende Kleinvögel bei Bodennebel. Journal für Ornithologie 137: 425-434.

Jenni, L. & Jenni-Eiermann, S. 1996. Metabolic responses to diurnal feeding patterns during the postbreeding, moulting and migratory periods in passerine birds. Functional Ecology 10: 73-80.

Jenni, L. & Widmer, F. 1996. Habitatnutzung von Kleinvögeln in der Herbstzugzeit am Neuenburgersee. Der Ornithologische Beobachter 93:221-248.

Jenni-Eiermann, S. & Jenni, L. 1991. Metabolic responses to flight and fasting in night migrating passerines. Journal of Comparative Physiology B 161: 465-474.

Jenni-Eiermann, S. & Jenni, L. 1994. Plasma metabolite levels predict individual body-mass changes in a small long-distance migrant, the Garden Warbler. Auk 111: 888-899.

Jenni-Eiermann, S. & Jenni, L. 1996. Metabolic differences between the postbreeding, moulting and migratory periods in feeding and fasting passerine birds. Functional Ecology 10: 62-72.

Kaiser, A. 1995. Estimating turnover, movements and capture parameters of resting passerines in standardized capture-recapture studies. Journal of Applied Statistics 22: 1039-1047.

Kaiser, A. 1996. Zugdisposition mitteleuropäischer Kleinvögel: Mauser, Körpermasse, Fettdeposition und Verweildauer. Journal für Ornithologie 137: 141-180.

Kirkwood, J.K. 1983. A limit to metabolisable energy intake in mammals and birds. Comparative Biochemistry and Physiology A 75: 1-3.

Klaassen, M. & Biebach, H. 1994. Energetics of fattening and starvation in the long-distance migratory garden warbler, Sylvia borin, during the migratory phase. Journal of Comparative Physiology B: 164: 362-371.

Klaassen, M., Lindström, Å. & Zijlstra, R. 1997. Composition of fuel stores and digestive limitations to fuel deposition rate in the long-distance migratory trush nightingale, Luscinia luscinia. Physiological Zoology 70: 125-133.

Kuenzi, A.J., Moore, F.R. & Simons, T.R. 1991. Stopover of neotropical landbird migrants on East Ship Island following trans-Gulf migration. Condor 93: 869-883.

Lavee, D., Safriel, U.N. & Meilijson, I. 1991. For how long do trans-Saharan migrants stop over at an oasis? Ornis Scandinavica 22: 33-44.

Leisler, B. 1990. Selection and use of habitat of wintering migrants. In: Gwinner, E. (ed) Bird migration; Berlin; Springer: 156-174.

Leisler, B. 1992. Habitat selection and coexistence of migrants and Afrotropical residents. Ibis 134 suppl. 1: 77-82.

Liechti, F., Steuri, T., Lopez-Jurado, C., Ribas, P.L.D., Reis, M.A. & Bruderer, B. 1997. Nocturnal spring migration on Mallorca - schedules of departure and passage. Ardeola 44: 207-213.

Lindström, Å. 1990. The role of predation risk in stopover habitat selection in migrating bramblings, Fringilla montifringilla. Behavioral Ecology 1:102-106.

Lindström, Å. 1991. Maximum fat deposition rates in migrating birds. Ornis Scandinavica 22: 12-19.

Lindström, Å. 1995. Stopover ecology of migrating birds: some unsolved questions. Israel Journal of Zoology 41: 407-416.

Lindström, Å., Hasselquist, D., Bensch, S. & Grahn, M. 1990. Asymmetric contests over resources for survival and migration: a field experiment with bluethroats. Animal Behaviour 40: 453-461.

Mädlow, W. 1994. Die Habitatwahl auf dem Wegzug rastender Kleinvögel in einer norddeutschen Uferzone. Acta ornithoecologica 3: 57-72.

Moore, F. & Kerlinger, P. 1987. Stopover and fat deposition by North American wood-warbler (Parulidae) following spring migration over the Gulf of Mexico. Oecologia 74: 47-54.

Morris, S.R. 1996. Mass loss and probability of stopover by migrant warblers during spring and fall migration. Journal of Field Ornithology 67: 456-462.

Morris, S.R., Richmond, M.E. & Holmes, D.W. 1994. Patterns of stopover by warblers during spring and fall migration on Appledore Island, Maine. Wilson Bulletin 106: 703-718.

Morris, S.R., Holmes, D.W. & Richmond, M.E. 1996. A ten-year study of the stopover patterns of migratory passerines during fall migration on Appledore Island, Maine. Condor 98: 395-409.

Pearson, D.J. & Lack, P.C. 1992. Migration patterns and habitat use by passerine and near-passerine migrant birds in eastern Africa. Ibis 134, suppl. 1: 89-98.

Piersma, T. & Gill Jr, R.E. 1998. Guts don’t fly: small digestive organs in obese Bar-tailed Godwits. Auk 115: 196-203.

Piersma, T. & Lindström, Å. 1997. Rapid reversible changes in organ size as a component of adaptive behaviour. Trends in Ecology and Evolution 12: 134-138.

Pilo, B. & George, J.C. 1983. Diurnal and seasonal variation in liver glycogen and fat in relation to metabolic status of liver and M. pectoralis in the Migratory Starling Sturnus roseus, wintering in India. Comparative Biochemistry and Physiology 74A: 601-604.

Pradel, R. 1996. Utilisation of capture-mark-recapture for the study of recruitment and population growth rate. Biometrics 52: 703-709.

Rappole, J.H. & Warner, D.W. 1976. Relationships between behavior, physiology and weather in avian transients at a migratory stopover site. Oecologia 26: 193-212.

Richardson, W.J. 1990. Timing of bird migration in relation to weather: updated review. In: E. Gwinner (ed.) Bird migration; Berlin; Springer: 78-101.

Safriel, U.N. & Lavee, D. 1988. Weight changes of cross-desert migrants at an oasis – do energetic considerations alone determine the length of stopover? Oecologia 76: 611-619.

Stephens, D.W. & Krebs, J.R. 1986. Foraging theory. Princeton; Princeton University Press.

Streif, M. 1991. Analyse der Biotoppräferenzen auf dem Wegzug in Süddeutschland rastender Kleinvögel. Ornithologische Jahreshefte Baden-Württemberg 7:1-132.

Sutter, E. 1957. Radar-Beobachtungen über den Verlauf des nächtlichen Vogelzuges. Revue Suisse de Zoologie 64: 294-303.

Weber, T.P., Houston, A.I. & Ens, B.J. 1994. Optimal departure fat loads and stopover site use in avian migration: an analytical model. Proceedings of the Royal Society, London B 258: 29-34.

Wingfield, J.C., Schwabl, H. & Mattocks Jr, P.W. 1990. Endocrine mechanisms of migration. In: Gwinner, E. (ed) Bird Migration; Berlin; Springer: 232-256.

Winker, K. 1995. Autumn stopover on the Isthmus of Tehuantepec by woodland nearctic-neotropic migrants. Auk 112: 690-700.

Winker, K., Warner, D.W. & Weisbrod, A.R. 1992. Daily mass gains among woodland migrants at an inland stopover site. Auk 109: 853-862.

Winkler, R. 1984. Avifauna der Schweiz, eine kommentierte Artenliste. Der Ornithologische Beobachter, Beiheft 5.

Woodrey, M.S. & Moore, F.R. 1997. Age-related differences in the stopover of fall landbird migrants on the coast of Alabama. Auk 114: 695-707.

Yong, W. & Moore, F.R. 1997. Spring stopover of intercontinental migratory trushes along the northern coast of the Gulf of Mexico. Auk 114: 263-278.

Young Jr, D.P. 1990. Resource defense by a migrating Song Sparrow. Journal of Field Ornithology 61: 232-236.