Statistical analysis
Statistical analyses were carried in JMP Pro 14.
Fertilization effects on plants: Specific leaf area (SLA), plant protein, and plant carbohydrates were analyzed with ANOVAs followed by a Tukey HSD when we found differences among treatments. To evaluate effects on ratio, plant protein and carbohydrate content was analyzed by MANOVA, we used a Pillai’s test statistic.
Choice experiment : the dry amount of plant eaten for each treatment was compared using MANCOVA techniques; we used Pillai’s test statistic. We used start mass as a covariate to correct for size differences among individuals. We included sex as an independent variable when its effect was significant. The dry amount of food eaten by males, and by females, was analyzed using ANOVAs followed by a Tukey HSD when we found differences among treatments.
Locust performance in the field: The number of locusts alive over time was analyzed by survival analysis followed by post-hoc comparison with Bonferroni correction (81). Total egg mass, egg number, and average egg mass were analyzed with ANCOVAs with pooled male and pooled female wet start masses as covariates and followed by a Student’s T test when we found differences among treatment.