Figure 1.

Changes in probability and intensity of the infection during the course of the experiment. (a) The probability of infection gradually decreased after the first week and by 30 days post-inoculation, most newts tested negative for Bd (GLMM: days post-inoculation; χ² = 30.12, df = 3, p < 0.0001; interaction sex x days post-inoculation χ² = 2.68, df = 3, p = 0.44).(b) The average infection burden was constant until 2 weeks after inoculation; during the third week after inoculation the infection intensity started to decline and generally reached zero at day 30 post-inoculation (GLMM: days post-inoculation; χ² = 9.20, df = 3, p < 0.001; interaction sex x days post-inoculation χ² = 0.01, df = 3, p = 0.99). Figures show predicted means ± SE generated from the mixed model analyses. In both analyses, tank had no influence on patterns of Bd infection (tank component = 0), but there was some consistency in the prevalence of Bd infection within versus among individuals (component = 0.01 ±0.009 (S.E.)) and a significant difference between individuals in Bd loads (component = 0.12 ±0.004 (S.E.)).