Data Analysis
All statistical analyses were done using SAS version 9.4 (SAS Institute
Inc., Cary, NC, USA). Data were log-transformed to linearize
relationships between dependent variables and SVL. We compared overall
growth rates between years and between sexes by using proc GLM to
regress growth rate versus the earliest SVL measurement of the year for
each lizard, with year and sex as covariates. We calculated body
condition separately in June and in July as the residual of log mass
regressed on log SVL using measurements taken in the first and last
weeks of the interval defined above. To analyze effects of mite load on
growth rate, we removed the effects of body size, sex, and year on
growth rate and mite load by computing residuals from separate
regression analyses of growth rate versus SVL and mite load versus SVL,
with sex and year as main effects. We then regressed residual variation
in growth rate or body condition versus residual variation in mite load.
We analyzed effects of residual mite loads on residual overall growth
rates separately for June and July because of the substantial difference
in environmental mite exposure between the two months and because mite
loads become higher on yearling males than on females in the first two
weeks of July (Pollock and John-Alder, 2020).