Lower temperatures decrease worker size variation but do not affect fine-grained thermoregulation in bumble bees

Abstract: Phenotypic variation within biological systems is ubiquitous and often assumed to be adaptive in social insects. Local environmental factors, such as temperature, may affect the phenotypic variation produced, and yield insights into the mechanisms that generate this group-level outcome. For instance, fine-scale heterogeneity in temperature across the nest may generate phenotypic variation by affecting larval development directly, or indirectly by inducing changes in the behavior of group members. To understand the role of temperature in producing phenotypic variation, we studied whether microclimatic differences within the nest could produce the worker size variation commonly observed in bumble bee (Bombus impatiens) colonies. We also tested if changes to the ambient temperature that colonies are exposed to influences the size variation they produced. We recorded brood temperature via thermal imaging and measured thorax widths of workers produced by colonies kept at ambient temperatures of 30 and 20 °C. Overall, average brood temperature did not differ across the nest even while the average temperature of non-brood structures (i.e., honeypots) interspersed among the brood decreased towards the periphery, demonstrating that bees were able to regulate brood temperature at a fine scale. However, we found that the size variation produced was sensitive to ambient temperature and increased under warmer temperatures. These results demonstrate that bumble bees have a unique method of fine-grained thermoregulation, where they regulate brood temperature but not the regions between brood. Additionally, while temperature is not the mechanism that produces size variation, it indirectly influences the mechanism that does. Significance statement: Morphological variation among group members affects group performance. We studied how the thermal environment experienced by developing workers, i.e., brood, influences the between-worker size variation produced within B. impatiens colonies. We found that temperature only indirectly influences size variation, as size variation increased under warmer nest temperatures, yet the temperature of brood did not differ across the nest. Our finding that workers maintain a uniform environment for the brood differs from previous results examining feeding rates across the nest and suggests that while the same workers tend to feed and thermoregulate the brood, they do so according to different rules and that by influencing worker behaviors, such as feeding rate, temperature affects size variation.