The demographic costs of nectar production in the desert perennial Prosopis glandulosa (Mimosoideae): a modular approach

Nectar production in angiosperms is considered to represent a reproductive cost, and has been associated with a decrease in fruit set or an overall decrease in the energetic budget of the plant. Populations of Prosopis glandulosa var. torreyana (honey mesquite) are a suitable system to evaluate the demographic costs of nectar production, as populations are composed of a 1:1 proportion of nectarful to nectarless individuals. The study was carried out in a population of 404 individuals of Prosopis g1andulosa var. torreyana found in an area with differing water availability in the Southern Chihuahuan Desert. The possible costs of nectar production were assessed on 1212 shoots of the honey mesquite that were tagged in 1994 and followed until 1998. We used two methods of analysis to describe the effect of nectar production on modular population dynamics: matrix analysis and log-xlinear models. Water availability and the varying environmental conditions affected plant growth, but nectar production did not have an effect on the demographic parameters we measured. The values of λ did not differ between nectar morphs and the only important effects we detected were the year to year variation in precipitation and microclimate differences at each site. Furthermore, the elasticity of each demographic process (growth, fecundity, retrogression and stasis) between nectar morphs did not differ. The log-linear models suggested a similar pattern but could discriminate the importance of each factor (nectar morph, year and site) on module fate. We were not able to detect a demographic cost of nectar production in the honey mesquite. The absence of a demographic response could be due to the negligible cost of producing nectar for this species or that the resources allocated for growth are different from those allocated for reproduction. Our results suggest that the modular fates of mesquites are mainly determined by environmental factors.