Experimental Demography of the Sand-Dune Annual, Cakile Edentula, Growing Along an Environmental Gradient in Nova Scotia

(1) A method of analysis is proposed to test whether differences in survivorship of a given species among two or more habitats are the result of variation among the habitats in environmental factors (density-independent effects), or in species density (densitydependent effects). (2) The method requires that a range of seed densities of the species be sown in a given habitat, and survivorship and reproductive output measured for each sowing density. If a straight line with negative slope can be fitted to the survivorship v. sowing density relationship, then the survivorship y-intercept of the line is an estimate of densityindependent mortality, and the slope is an estimate of the density dependence of mortality. (3) If this analysis is carried out for the species growing in several habitats, it is possible to test whether intercepts or slopes (and consequently density-independent or density-dependent effects) differ among habitats. A similar analysis can be carried out for reproductive output. (4) The annual plant Cakile edentula, a common species on sand dunes in eastern North America, is used as an example. Three experimental sites along a sand-dune gradient were each sown with seeds at a range of densities and survivorship and reproductive output measured for each density at each site. (5) The density dependence of survivorship changed significantly (P < 0.001) along the gradient (among habitats), whereas density-independent mortality stayed constant. (6) Significant differences (P < 0.001) were observed for both density-independent and density-dependent effects on reproductive output. (7) The effects of density on survivorship were greatest at the landward end of the gradient, whereas the effects of density on reproductive output were greatest at the seaward end.