Plasticity in phytolith production in Asteraceae species along a rainfall gradient in Israel

Background/Question/Methods

Phytoliths are microscopic bodies of amorphous silica deposited in association with cells in various plants. Members of the grass family (Poaceae) produce more phytoliths than most angiosperms, and are therefore more intensively studied. Studies on grasses have suggested that three main environmental factors promote silicification and phytolith production in plants: silica availability, water availability and transpiration. There is increasing evidence that water availability in arid regions is positively correlated with phytolith production by grasses. Studies have focused on the water availability variable in several grass species growing along relatively limited gradients (usually 200-250 mm per year). We have extended the range of available data by studying a larger rainfall gradient in Israel (80-450 mm mean annual rainfall) in which potential evaporation and soils also vary. Along this gradient we sampled annual plants from two non-spiny and one spiny Asteraceae species and an additional non-spiny grass species. All species were sampled in ungrazed sites in similar contexts. Phytoliths were extracted using dry ashing and quantified following Albert et al.'s (1999) procedure.

Results/Conclusions

One of the non-spiny Asteraceae species was sampled in one of the sites in two consecutive years, and phytolith concentrations were indeed higher in the wetter year. This observation demonstrates that isolation of the water availability variable produces the same result in grasses and the Asteraceae. However, inter-site analysis has indicated that for the grass species and the two non-spiny Asteraceae species, phytolith concentrations are negatively correlated with water availability. Since phytoliths are known to protect plants against both aridity and herbivory, under extremely arid conditions selection probably favors plants which produce more phytoliths. In contrast, the spiny species was found to produce more phytoliths in the least arid site, which we assumeto be related to the existence of a prominent physical antiherbivory defense (spines). Based on these results, we hypothesize that adaptation to extremely arid conditions plays a more important role than water availability in non-spiny species. The spiny species, which are protected by spines, possibly experiences less selection in favor of silicification as defense and thus an increased effect of water availability on silicification. Nonetheless, more data are needed to support these suggestions.