The lack of any phylogenetic influence on virulence suggests that
ecological reasons underlie the variation in virulence of the lizard malaria
parasites. Therefore, I now revisit the hypotheses presented previously to
determine if the present ecology of the lizard malaria parasites differs in
ways that would predict variation in virulence.Evaluation of hypotheses
Host perspective
No data are available on the cost to lizards of mounting an immune
response to malaria infection. This is not unusual, as data on this issue are
rare (Gemmill and Read, 1998). I have followed both captive and free-
ranging animals infected with thePlasmodiumspecies discussed here,
as well as lizards infected with various haemogregarines, microfilarial
worms and a trypanosome, and have rarely noted elimination of the
parasite. The only exception was a virus that forms huge assembly pools
within erythrocytes ofA. agama. These infections were acute and were
eliminated within a few weeks. This anecdotal evidence suggests that liz-
ards do not mount a particularly effective immune response to protozoan
blood parasites. This is surprising, because there is a substantial fitness
cost (reduction in clutch size) associated with infection byP. mexicanum
and mixed infections of the two African species.Transmission-opportunity hypothesis
Comparisons among the lizard malaria systems do not give support
for this hypothesis.P. mexicanumin fence lizards appears to be the
most virulent of the seven species surveyed here and yet transmission
is strongly seasonal. Overall prevalence of P. mexicanum at the
California site is somewhat lower than for the tropical systems, suggesting
that transmission is less intense. For the two tropical systems, vector
biting activity may be more uniform in the Caribbean, where there
is no clear dry season with reduced transmission. However, the three
Caribbean lizard malaria species appear to be least harmful to their
host.Mobility hypothesis and transmission-mode hypothesis
These hypotheses should be irrelevant for comparisons among lizard
malaria, as all are transmitted by blood-feeding Diptera.Host-demography hypothesis
Mark–recapture programmes at all sites reveal that anoles, fence lizards
and rainbow lizards can all live for several years. However, rainbow
lizards are much larger animals and would be expected to have longer
lifespans, and fence lizards suffer heavy winter mortality in California.
This suggests that the virulence of malaria in fence lizards should be theParasite Virulence 305