JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Native forest species exhibit a well‐known range of ecological roles with respect to natural disturbance regimes, from pioneer phase to mature phase, and they regenerate from a range of sources, including dormant seeds, seed rain, pre‐established juveniles, and resprouts from damaged adults. In contrast, the ecological roles of invasive, non‐indigenous species in forest communities after natural disturbances are not well understood. Some previous studies of invasive species have emphasized their weedy nature and their ability to colonize anthropogenic disturbances. Tropical hardwood hammock forests in southern Florida experience frequent disturbance by hurricanes. Our studies of forest regeneration during two years following a recent severe hurricane suggest that invasive non‐indigenous forest species exhibit the same range of ecological roles as native forest species and compete with native species for particular kinds of regeneration opportunities. To study ecological roles of non‐indigenous species in regenerating forests after Hurricane Andrew, we set up four large study areas at each of three study sites that had differing amounts of hurricane‐caused canopy disturbance. There were two pairs of 30 × 60 m research plots per site, and in each pair there was one control plot and one restoration plot; restoration areas were subject to an aggressive management program, focused on reducing non‐indigenous vine cover. Within these study areas we subsampled vegetation in small study plots that were regularly spaced, and conducted vegetation censuses in April (the end of the dry season) and October (the end of the rainy season) for 2 yr, beginning in April 1993. We found that the source of regeneration for forest species was dependent upon the amount of canopy disturbance, the time since disturbance, and the autecology of the constituent species. Overall, 28% of the 90 species were non‐indigenous: 34% of the vines (N = 32) and 24% of other life‐forms (N = 58). Non‐indigenous vines seemed to have a special role; not only could they compete with native vines, but they could also negatively affect the regeneration of other natives from a diverse array of sources including pre‐established juveniles and resprouts from damaged adults. Both native and non‐indigenous vine cover in unmanipulated study areas increased following the hurricane. Non‐indigenous vine species had higher cover than native vine species, and many species formed dense “blankets.” Non‐indigenous species in general (not just vines) did not differ significantly from native species in seed mass, nor were they restricted to the pioneer type of life history. Many non‐indigenous species had invaded forests prior to hurricane disturbance and had their own banks of pre‐established juveniles; others recruited from dormant seeds, seed rain, and/or resprouts from pre‐established adults. Based on information on source of regeneration and impact on native species, we propose a classification scheme for functional roles of non‐indigenous invasive species in forests. To i...

The objective of the present study was to determine the plasma levels of testosterone and corticosterone in male Anolis sagrei during the annual reproductive cycle and to examine the relationships between seasonal change in the levels of these hormones, male reproductive activity, and body condition. Both testosterone and corticosterone levels in adult males captured in the Miami, Florida, area varied significantly with month and with season (i.e., breeding vs. nonbreeding period), although they were not significantly correlated with each other or with body mass. Mean monthly testosterone levels were higher during the breeding season (March-August) than during the nonbreeding season, with the one exception that the highest mean testosterone level occurred in February immediately before the beginning of the breeding season. Testosterone levels in the overall sample of 144 males were significantly correlated with testes mass but not with any of the other measured variables. Corticosterone levels were highest during the nonbreeding season and lowest during the breeding season. Corticosterone levels on a monthly basis were negatively correlated with monthly changes in testes mass and positively correlated with monthly changes in abdominal fat-body mass.

No abstract