Recent theoretical developments concerning the evolution of species interactions in general and mutualism in particular, have emphasized the need to consider the geographical, environmental and spatial contexts within which species interactions occur and evolve.
Myrmecochory, seed dispersal by ants, is a mutualistic animal-plant interaction whose ecology and evolution have been thoroughly studied in several ecosystems around the world.  However, although some plant species in Mediterranean and desert ecosystems exhibit morphological adaptations that characterize dispersal by ants, our knowledge of that type of interactions in the abiotic and biotic contexts that such ecosystems provide is very poor.
The main goal of this project is to test whether ideas and hypotheses that were generated, tested and supported in regions where myrmecochory is well studied, are valid and gain support in those (semi-)arid ecosystems where the payoffs for the participants (plants and ants) are context-dependent and less certain.
a.     Seed dispersal of Sternbergia clusiana
We are looking at the ecological significance and selection for seed dispersal by ants along a sharp environmental gradient.  Our model system is the plant Sternbergia clusiana and the ants with which it interacts. This rare plant, which is of conservation, cultural and aesthetic value, has a very patchy distribution that may reflect strict availability of suitable habitat and/or very limited seed dispersal capabilities. This plant has many characteristics and adaptations that make it a very good candidate for being dispersed by ants. We hypothesize that the plant adaptations for dispersal by ants evolved in a different ecological and evolutionary context. We use a combination of observations, field experiments, chemical analyses to test whether the investment of the plant in dispersal adaptations (i.e. investment in reward for the ants), the distribution of seed-dispersing ants and the behavioral responses of the ants vary with geographic position and biotic context. So far, we found that a) there is a geographical gradient in some aspects of the investment (elaiosome size), but not of others (elaiosome chemistry), this investment is correlated with the behavioral responses by different guilds of potentially seed-dispersing ants; especially when compare to co-occurring myrmecochores. We also found that ‘classical’ seed-dispersing guilds (scavenging ants) provide benefits in phases that follow the initial dispersal. This plant is a representative of a small group of plants in the Israeli flora, most of which are rare, that share similar adaptation for dispersal by ants. We now expand our research to other species in the same study region for a broader comparison across many taxa.