Abstract
The integration of rapidly renewable biomass into insect production systems has been proposed as a strategy to improve resource-use efficiency in insect production. This study evaluated the graded inclusion levels of dried watermeal (Wolffia globosa) in diets of two-spotted crickets (Gryllus bimaculatus) and assessed voluntary nutrient regulation under free-choice feeding. Four fixed-inclusion diets (0%, 25%, 35%, and 45% watermeal) and one self-selection treatment were tested over 28 days. Growth performance, feed conversion ratio (FCR), survival rate (Surv), production index (PI), and whole-body composition were determined. Repeated-measures analysis using linear mixed-effects models indicated that treatment, week, and their interaction were statistically significant (p ≤ 0.024). However, partial R(2) analysis showed that the independent contributions of treatment and week were negligible, whereas the treatment × week interaction explained measurable variance, indicating that dietary effects were primarily expressed through time-dependent responses. Segmented regression identified a breakpoint at 35% watermeal inclusion (95% CI: 24.93-45.07), indicating that PI was the highest within a moderate supplementation range under the present fixed-diet conditions rather than at a precise single optimum. Inclusion levels beyond this threshold reduced performance. Under free-choice conditions, crickets progressively increased watermeal intake with age and maintained stable nitrogen-free extract (NFE):crude protein (CP) and gross energy (GE):CP intake ratios, selecting an average of 25-35% watermeal over the experimental period. This supplementation range improved feed efficiency and protein deposition while limiting lipid accumulation, suggesting improved energy-protein balance and nutrient partitioning. The self-selection result is interpreted as evidence of behavioral intake regulation under choice conditions and not as direct validation of the segmented-regression breakpoint. Collectively, these findings provide complementary statistical and behavioral evidence supporting a biologically relevant moderate inclusion range (approximately 30-35%) of dried watermeal.