Plants interact with a diversity of microbial symbionts throughout their life cycle. However, these interactions are not static throughout the lifetime of the host plant, and it is possible that a symbiont may have positive or negative effects on its host depending on the life stage. How do these combinations of costs and benefits balance out to alter host population growth? In addition, how do differences in host life history affect their demographic response to symbiosis? We address these questions by focusing on the interaction between a pair congener grass hosts that differ in regional abundance patterns and their vertically-transmitted fungal endophytes. Using 5 years of demographic data from experimental populations, we explore the effects of endophyte symbiosis on host population growth using integral projection models (IPM). We compare the effects of endophytes on the two hosts for each demographic transition, as well as their contributions to the difference in lambda in endophyte-free compared to endophyte-infected populations. Finally, we use a megamatrix approach to incorporate the effects of imperfect transmission on endophyte persistence in an infected population as well as host population-level responses.
Results/Conclusions
We found that overall, endophyte symbiosis increased host population growth. However, endophyte symbiosis had differing effects on the demographic transitions of the two host species. In Poa sylvestris (POSY), the more common host, endophytes increased lambda through positive effects on growth and survival, with modest positive contributions through reproduction. Whereas for Poa alsodes (POAL), the rare host, endophytes increased lambda through positive effects on survival and seedling establishment, with costs to size-independent seed production. POSY populations were projected to increase (lambda > 1) regardless of endophyte status, whereas POAL populations could only increase under endophyte symbiosis. Vertical transmission rates interacted with probabilities of seedling establishment in complex ways to alter host population growth and the stable persistence of endophytes in the host population. We saw evidence of “threshold transmission rates” under which endophyte symbiosis had no effect on the host population. POSY populations exhibited lower threshold transmission rates compared to POAL populations, indicating a lower threshold to the benefits of endophyte symbiosis. Our results demonstrate that complex interactions between host and symbiont life histories combine to affect host population dynamics, and highlight the importance of investigating host-symbiont interactions not just at particular life stages, but at the population level.