As bumble bees may routinely forage on pesticide-treated plants, it is imperative to apply pesticides that pose minimal risk. Identifying such reduced-risk pesticides requires toxicity tests using biologically relevant and standardized exposure methods. Our goal was to develop a functional and reliable queen-less micro-colony method for determining the sublethal effects of pesticides on Bombus impatiens Cresson, a species for which further toxicity assessments are needed. Originally developed using Bombus terrestris (L.), this method involves isolating three workers and allowing them to establish and maintain a colony. The workers and brood can then be exposed to pesticides and various effects measured simultaneously. Our preliminary work indicated that B. impatiens workers were unable to reproduce under the micro-colony conditions developed for B. terrestris. In the current study, we focused on the potential role of group size in B. impatiens worker reproduction by placing groups of three or five sister workers together and assessing their ability to establish a colony and rear males to adulthood. While both groups oviposited and produced males, significantly more males emerged from micro-colonies of five workers. We conclude that group size affects B. impatiens worker reproduction, and this effect in bumble bees varies interspecifically. In the context of toxicity testing, we recommend the use of at least five B. impatiens workers in micro-colonies to ensure a level of reproduction that allows the detection of potential sublethal changes to reproduction or development.