Abstract. The geographic mosaic theory of coevolution posits that the form of selection between interacting species varies across a landscape with coevolution important and active in some locations (i.e., coevolutionary hotspots) but not in others (i.e., coevolutionary coldspots). We tested the hypothesis that the presence of red squirrels (Tamiasciurus hudsonicus) affects the occurrence of coevolution between red crossbills (Loxia curvirostra complex) and Rocky Mountain lodgepole pine (Pinus contorta ssp. latifolia) and thereby provides a mechanism giving rise to a geographic mosaic of selection. Red squirrels are the predominant predispersal seed predator and selective agent on lodgepole pine cones. However, in four isolated mountain ranges east and west of the Rocky Mountains, red squirrels are absent and red crossbills are the main predispersal seed predator. These isolated populations of pine have apparently evolved without Tamiasciurus for about 10,000 to 12,000 years. Based on published morphological, genetic, and paleobotanical studies, we infer that cone traits in these isolated populations that show parallel differences from cones in the Rocky Mountains have changed in parallel. We used data on crossbill and conifer cone morphology and feeding preferences and efficiency to detect whether red crossbills and lodgepole pine exhibit reciprocal adaptations, which would imply coevolution. Cone traits that act to deter Tamiasciurus and result in high ratios of cone mass to seed mass were less developed in the isolated populations. Cone traits that act to deter crossbills include larger and thicker scales and perhaps increased overlap between successive scales and were enhanced in the isolated populations. In the larger, isolated mountain ranges crossbills have evolved deeper, shorter, and therefore more decurved bills to exploit these cones. This provides crossbills with higher feeding rates, and the change in bill shape has improved efficiency by reducing the concomitant increases in body mass and daily energy expenditures that would have resulted if only bill size had increased. These parallel adaptations and counter adaptations in red crossbills and lodgepole pine are interpreted as reciprocal adaptations and imply that these crossbills and pine are in coevolutionary arms races where red squirrels are absent (i.e., coevolutionary hotspots) but not where red squirrels are present (i.e., coevolutionary coldspots). The view that coevolution is a prominent evolutionary process has been contentious. Nonetheless, evidence from diverse areas of evolutionary biology indicate that coevolution is a fundamental process in evolution and that a better understanding of the coevolutionary process is critical for progress in evolutionary biology, conservation, and health-related issues (Thompson 1994(Thompson , 1999a. One recent advance in coevolutionary theory has risen from the recognition that the form of an interaction between species often varies across a landscape (e.g., Thompson and Pellmyr 1992). This geographic v...

No abstract

No abstract

Although the habitat requirements of breeding populations of Henslow's Sparrow (Ammodramus henslowii) have been examined, less is known about their habitat requirements and ecology during the nonbreeding season. We estimated population densities and quantified habitat associations of Henslow's Sparrows wintering in saline soil barrens in southern Arkansas. Densities of Henslow's Sparrows in the saline soil barrens were similar to those in the Longleaf Pine (Pinus palustris) Ecosystem of the southeastern United States, considered by many to be their primary wintering habitat. Henslow's Sparrows were closely associated with open areas with greater cover of Aristida spp. and globe beaksedge (Rhynchospora globularis), greater stem density at 11–20 cm above ground, more lichens, more herbaceous cover, more bare ground, greater occurrence of little bluestem (Schizacyrium scoparium) as the tallest vegetation, less moss, and less shrub cover than randomly selected sites. In contrast to the results of studies conducted in the Longleaf Pine Ecosystem, the presence of Henslow's Sparrows in our study was not correlated with the height of the tallest vegetation. Our results indicate that saline soil barrens of southern Arkansas support a high density of wintering Henslow's Sparrows and do so for longer postdisturbance periods than longleaf pine savanna. We also found that stem density near the ground was similar to that reported from longleaf pine savanna, but only about half that observed on their breeding grounds. Areas used by Henslow's Sparrows had more lichen and less moss cover, suggesting that those areas were drier than random sites within the barrens. Further research is needed to determine if large populations of Henslow's Sparrows winter in other saline soil barrens and if fire influences habitat associations and densities in the barrens.

Smith's Longspurs (Calcarius pictus) are a species of concern in North America because of their limited range and apparent low population size. To better understand winter habitat needs and guide management of this species, we examined habitat associations of Smith's Longspurs in eastern Arkansas by comparing grassland patches where Smith's Longspurs flushed to randomly located patches in the same area. Smith's Longspurs were found in sparse grassland patches of relatively low height adjacent to airport runways where the native grass prairie three‐awn (Aristida oligantha) dominated ground cover and vertical structure. Smith's Longspurs were not found in vegetation plots dominated by non‐native Bermuda grass (Cynodon dactylon). Prairie three‐awn grass may provide concealment from predators and their seeds may be an important food source. Occurrence of Smith's Longspurs was also correlated with less litter, perhaps because deeper litter could make walking and searching for seeds more difficult. Availability of suitable habitat for Smith's Longspurs along airport runways may be declining due to natural succession of grassland habitat in the absence of disturbance and recent changes in management that favor Bermuda grass. Conversion and degradation of native prairie has resulted in the decline in abundance and distribution of Chestnut‐collared Longspurs (C. ornatus) and McCown's Longspurs (Rhynchophanes mccownii). Our findings suggest that conversion of native grasslands to non‐native grasslands results in loss and degradation of habitat for wintering populations of Smith's Longspurs.