Parasite resistance due to host habitat specialization in damselfly-water mite associations

Background/Question/Methods

There are competing hypotheses as to whether more widespread and abundant host species will have higher measures of parasitism.  The geographic range hypothesis states that the more widespread species will be more parasitized than its less widespread species.  The alternative hypothesis is that the more widespread species will use the parasites against its geographically restricted sibling species as a tool for successful competition.  Few study the resistance differences between conspecific hosts with different range sizes, regional and local abundances. The objective of this project was to compare two sympatric conspecific damselfly host species response to a generalist parasite water mite species. I aerial sweep net collected two host species, Nehalennia irene and Nehalennia gracilis at two bogs over a seven week period. N. irene is a geographically and regionally widespread species while N. gracilis is a bog specialist with a restricted geographic range and usually lower local abundance than N. irene. I assessed water mite parasitism, which identified to morphospecies as Arrenurus sp.  I then determined whether there are differences between levels of infection between the species using fisher exact test for parasite prevalence, standard bootstrapping for mean intensity of infection and Mood’s median test for median intensity.

Results/Conclusions

A total of 1007 damselfly individuals were collected for this study. There was no sex bias in parasitism for either species at either site. Because we collected over a seven week period, we checked whether we can pool the prevalence data over the seven weeks. There was no significant difference in prevalence of infection between the seven weeks; therefore the seven weeks were pooled as well.  Prevalence of water mite infection between the two host species was not significantly different in either Hebert Bog (P=0.19) or Westport Bog (P=0.124). Mean and median intensity of infection also did not vary between the species in either Hebert Bog (t=0.327 P=0.763 for mean intensity and P=0.446 for median intensity) or Westport Bog (t= - 0.746 P=0.478 for mean intensity and P=0.288 for mean intensity).  However all the water mites were resisted in N. gracilis but none are resisted in N. irene. Resistance therefore is occurring in the more geographically restricted host supporting the geographic range hypothesis. However, this species is also more locally adapted or specialised in its habitat which may allow it to be better adapted to resist the parasites in that area.