ABSTRACT Hummingbird migration has long fascinated researchers due to the limitations of small body size and high metabolic rate on migratory performance. Yet, few data are available concerning hummingbird migration strategies, especially for species that must negotiate major geographic barriers. To address this problem, we investigated the migration ecology of Ruby-throated Hummingbirds (Archilochus colubris) passing through a coastal banding station in southwest Louisiana following flights across the Gulf of Mexico. Our aims were to describe the phenology of spring migration and explore potential differences in the migration ecology of males and females. During our 10-yr study (1998–2007), we found that peak hummingbird passage generally occurred in the second half of April and that males preceded females by about three days. However, females arrived in significantly better energetic condition as measured by fat and muscle stores as well as size-corrected body mass. Most birds did not stay at our study site to refuel, only 2% of individually marked birds were recaptured more than a day after initial capture (range = 1–5 d). Our results suggest that Ruby-throated Hummingbirds exhibit protandrous migration (i.e., males migrate earlier) and that en route body condition may be a consequence of sexual dimorphism in wing morphology (i.e., lower wing loading in females) that allows females to expend less energy during migration across the Gulf of Mexico.
La migración de colibríes siempre ha fascinado a investigadores debido a las limitaciones de tamaño pequeño de cuerpo y alta tasa metabólica en el rendimiento migratorio. Sin embargo, hay pocos datos disponibles sobre las estrategias de migración de colibríes, especialmente para las especies que deben negociar grandes barreras geográficas. Nosotros investigamos la ecología de migración del Colibrí Garganta Rubí (Archilochus colubris) que pasa a través de una estación de anillamiento en la costa suroeste de Louisiana después de sus vuelos a través del Golfo de México. Nuestros objetivos fueron describir la fenología de la migración de primavera y explorar las posibles diferencias en la ecología de la migración de machos y hembras. Durante nuestros 10 años de estudio (1998–2007), encontramos que el pico de paso de colibrí generalmente ocurrió en la segunda quincena de abril y que los machos precedieron a las hembras por unos tres días. Sin embargo, las hembras llegaron en significativa y mejor condición energética medido por medio de reservas de grasa y músculo, así como el tamaño corregido por masa corporal. La mayoría de las aves no se quedaron en nuestro sitio de estudio para re-abastecerse, sólo el 2% de las aves marcadas individualmente fueron capturados más de un día después de la captura inicial (rango = 1–5 días). Nuestros resultados sugieren que los Colibríes Garganta Rubí exhiben una migración protandra (es decir, los machos migran antes) y que la condición corporal en-ruta puede ser una consecuencia del dimorfismo sexual en la morfología del ala (es decir, menor carga alar en las hembras) que permite a las hembras de gastar menos energía durante la migración a través del Golfo de México.
Geographic barriers, like the Gulf of Mexico, magnify the energetic requirement of migration because many migrant species cross these barriers in a single long-distance flight (Lowery 1946). Because males and females are thought to experience different selective pressures with respect to timing of migration and arrival at the breeding grounds (Morbey and Ydenberg 2001, Dierschke et al. 2005), they may respond to the energetic challenge of crossing a barrier in different ways. For example, in protandrous migrants, males may be under stronger time selection, and therefore, may fuel faster and deposit larger fat stores than females before barrier crossing (Alerstam and Lindström 1990, Dierschke et al. 2005). However, few studies have investigated this possibility. Among trans-Gulf migrants, with the exception of male Magnolia Warblers (Dendroica magnolia) and Tennessee Warblers (Vermivora peregrina) that arrive with larger fat stores than females, males do not seem to differ from females in the amount of fat they carry upon arrival at the northern coast of the Gulf of Mexico (Mabey 2002).
Ruby-throated Hummingbirds (Archilochus colubris) are the only hummingbird species in eastern North America and the only one that crosses a major geographic barrier (the Gulf of Mexico) in a single flight during its long-distance migration (Robinson et al. 1996). Considering the conflict of high energetic demand of migration and small body size that limits fat deposition as well as their high metabolic rate, Ruby-throated Hummingbirds offer an interesting system for studying adaptations to long-distance migration. To improve our understanding of their migratory behavior, we analyzed capture data for migrating Ruby-throated Hummingbirds following the crossing of an ecological barrier, the Gulf of Mexico. Our specific objectives were to (1) describe the phenology of spring migration by Ruby-throated Hummingbirds through a stopover site along the northern coast of the Gulf of Mexico, and (2) investigate possible differences in the migration patterns and energetic conditions of males and females.
Data were collected at a banding station (29°45′ N, 93°37′ W) near Johnson Bayou in southwestern Louisiana. Hummingbirds were captured using 28 to 30 mist nets (12 × 2.6 m; 30-mm mesh) as part of a long-term banding operation in a coastal forested wetland site (see Barrow et al. 2000 for habitat description). Netting was conducted daily from 7:00 to 16:00 CST from late March until early May 1998 to 2007. As part of our protocol, we collected feather samples (rectrices) from all transient migratory species. However, due to concerns about impaired flight performance and maneuverability (Chai et al. 1999), we only clipped the tip (∼2–3 mm) of one of the outer rectrices. We did not band hummingbirds, so clipping allowed us to identify recaptures. During 2002 and from 2004 to 2007, hummingbirds (N= 551) received a unique clipping pattern that permitted individual identification if recaptured within the same season, allowing us to gather information about recapture rates, mass change, and minimum stopover duration. We determined the sex and measured unflattened wing length (± 0.5 mm) and body mass (± 0.1 g) of all captured individuals. We quantified fat stores by scoring visible subcutaneous fat deposits in the furcular and abdominal regions using a six-point ordinal scale, with 0 indicating a lack of visible fat stores and 5 indicating bulging fat deposits in the furcular and abdominal regions (Helms and Drury 1960). Pectoral muscle mass was scored on a four-point ordinal scale based on the thickness of the major flight muscle (pectoralis major; Bairlein 1995), with 0 indicating a low concavely shaped muscle with a protruding “sharp” sternum and 3 indicating a bulging convexly shaped muscle.
Data analyses We used the residuals from a regression of body mass on wing chord as a size-corrected measure of body condition. We estimated minimum stopover duration by subtracting the date of first capture from the last date of recapture (Cherry 1982) for birds recaptured at least a day after the first capture. Mann-Whitney U-tests were used to compare passage dates within and across years between the sexes. Arrival body condition of the sexes was also compared using a Mann-Whitney U-test. Chi-square tests were used to compare the fat and muscle scores of males and females. Statistical analyses were performed using SPSS 15.0 (SPSS 2006) and all tests were two-tailed.
We captured 956 Ruby-throated Hummingbirds (Nmales= 425, Nfemales= 531) during our 10-yr study (Table 1). Overall, the median passage date of males was three days earlier than that of females (in nonleap years: 19 April vs. 22 April; U= 94 558, P < 0.0001), and passage dates ranged from 20 March to 7 May for males and 26 March to 7 May for females (Fig. 1). Although the median arrival date of males preceded that of females during each year of our study, the difference was significant only in 2001 (Table 1). In general, females arrived in better body condition as indicated by larger size-corrected body mass (U= 80 709.5, P= 0.021), larger fat stores (χ21= 16.2, P= 0.006; Fig. 2A), and greater pectoral muscle mass (χ21= 17.2, P= 0.001; Fig. 2B).
Table 1. Year of study, capture effort, and comparison of median passage times of male and female Ruby-throated Hummingbirds during spring migration in southwestern coastal Louisiana.
Individuals captured (median passage date)
Dates of operation
Comparison of passage datesa(P)
28 March–7 May
15 (10 April)
22 (21 April)
27 March–7 May
19 (06 April)
11 (27 April)
27 March–7 May
52 (18 April)
56 (19 April)
26 March–6 May
26 (22 April)
59 (25 April)
25 March–5 May
59 (20 April)
58 (23 April)
25 March–7 May
62 (22 April)
83 (23 April)
24 March–7 May
39 (18 April)
38 (21 April)
23 March–6 May
44 (19 April)
36 (20 April)
24 March–6 May
34 (8 April)
38 (19 April)
18 March–6 May
75 (13 April)
130 (14 April)
We recaptured 2.0% (11 of 551, Nmales= 4, Nfemales= 7) of individually marked hummingbirds after the day of first capture in 2002 and from 2004 to 2007. Median minimum stopover length was 3.5 d for males (range = 1–5 d) and 2 d for females (range = 1–5 d). Mean mass change (±SD) was 0.35 ± 0.55 g for males, a 12.6 ± 19.4% increase over the mean initial capture mass, and 0.03 ± 0.5 g for females, a 1.2 ± 15.2% increase over the mean initial capture mass.
Peak hummingbird passage in coastal southwestern Louisiana generally occurred in the second half of April, with male Ruby-throated Hummingbirds preceding females by an average of three days. Because the first males may arrive at the northern coast of the Gulf of Mexico as early as late February (Robinson et al. 1996), our results were likely influenced by the limits of our sampling period. The difference between passage dates of the earliest male and female migrants was six days, consistent with data reported previously. The first male Ruby-throated Hummingbird passed through Horn Island, a barrier island ∼10 km off the Mississippi coast, five days earlier than the first female (Moore et al. 1990), whereas this difference was 8–10 d on the Alabama Gulf coast (Robinson et al. 1996) and at least 10 d in Pennsylvania (Mulvihill and Leberman 1987). Similarly, for all migratory hummingbird species in western North America, males are reported to arrive in breeding areas earlier than females, although quantitative studies of migration timing and ecology are lacking (Calder and Calder 1992, 1994, Johnsgard 1997, Calder 1999, Baltosser and Russell 2000, Healy and Calder 2006).
We found that female Ruby-throated Hummingbirds reached our study site in better energetic condition than males. Because the energetic cost of flight varies with wing loading (Pennycuick 1975, Norberg 1990), males with shorter wings and higher wing loading (i.e., body mass relative to wing area) should have greater energetic costs when crossing a large barrier such as the Gulf of Mexico (Chai and Dudley 1999). Feinsinger and Chaplin (1975) found that species or sexes of hummingbirds that were territorial had higher wing loading than nonterritorial birds, suggesting a trade-off between energetically efficient flight and aerial display/fighting abilities (Kodric-Brown and Brown 1978, Chai and Dudley 1999). This sex-based difference in flight energetics also leads to different resource-exploitation strategies (Feinsinger and Chaplin 1975, Kodric-Brown and Brown 1978). Due to their higher competitive and fighting abilities (facilitated by shorter, more pointed wings), males of most territorial hummingbird species, including Ruby-throated Hummingbirds, are dominant over females and monopolize high-quality nectar patches, with females excluded to lower quality foraging areas (Pitelka 1942, Johnsgard 1997). For example, anecdotal evidence suggests that male Ruby-throated Hummingbirds monopolize higher quality nectar sources during spring stopover on Horn Island, a barrier island off the coast of Mississippi, with females found in lower quality habitats (Osborne 1998). Females move into previously occupied higher quality habitats only after males leave (Osborne 1998). However, because of the morphological differences, females may expend less energy during foraging (Kodric-Brown and Brown 1978, Carpenter et al. 1993). This energetic compensation could be critical during migration when rapid refueling is necessary for timely arrival in breeding areas. For example, territorial male and female Rufous Hummingbirds (Selasphorus rufus) had similar refueling rates and stopover lengths during migration, possibly because females compensate for their lower quality territories with more energetically efficient foraging (Carpenter et al. 1991, 1993). However, among birds unable to obtain territories and restricted to exploiting distant, sparse nectar patches, females gained mass at a faster rate, likely due to lower flight costs during foraging (Carpenter et al. 1993).
Because of the small number of recaptured individuals (N= 11), we cannot draw meaningful inferences about the stopover strategies of male and female Ruby-throated Hummingbirds. Our limited data on mass change and minimum stopover length suggest that males may occupy higher quality nectar patches to refuel, but further study is needed. Our low recapture rate (2%) suggests that most Ruby-throated Hummingbirds did not spend more than a day at our coastal stopover site in Louisiana. Brief, less than a day, stopover durations by Ruby-throated Hummingbirds have also been reported in coastal Alabama (Robinson et al. 1996) and in Minnesota (Winker et al. 1992). Similarly, short spring stopovers and trap-lining-style en route foraging were reported for migrating Broad-tailed Hummingbirds (Selasphorus platycercus; Calder and Calder 1992). In contrast, studies of migrating Rufous Hummingbirds in the fall in the Sierra Nevada revealed that individuals defended territories and remained at stopover sites for one to two weeks before resuming migration (Carpenter et al. 1991, 1993). During the fall, migrating Ruby-throated Hummingbirds appear to employ a strategy based on short daily flights and brief stopovers (Willimont et al. 1988). This migratory strategy is consistent with our results and is similar to the fly-and-forage migration strategy employed by other daytime migrants (Strandberg and Alerstam 2007). Because hummingbird migration routes are mostly driven by the phenology and availability of food resources across the landscape (Phillips 1975, Johnsgard 1997), migration strategies likely differ among biogeographic regions as well as seasons. To elucidate these differences, more focused studies are necessary.
In conclusion, we found that male and female Ruby-throated Hummingbirds differ in the timing of spring passage through a stopover site along the northern coast of the Gulf of Mexico. Males arrived earlier and with less energy reserves than females. Males may be constrained by their shorter wings and higher wing loading, thus higher energy expenditure during trans-Gulf flight might result in lower body condition at arrival. However, their social dominance over females may allow males to acquire better quality resources and refuel at higher rates. Because differential selection pressures acting on the sexes may result in different migration strategies, additional studies of intersexual differences in stopover ecology and resource use are needed.
Our study was supported by grants from the National Science Foundation (IOB 0078189, DEB 05544754) to FRM. We are grateful to J. C. Owen, J. Mollenhauer, and numerous field technicians for their assistance in data collection over the years. The Gray Estate granted permission to conduct research on their property. Earlier drafts of the manuscript were improved due to valuable comments from R. J. Smith, G. Ritchison, and three anonymous reviewers. All procedures involving animals were in accordance with the Institutional Animal Care and Use Committee of the University of Southern Mississippi.