Figure 1. Experimental set-up of (BioBest) Bombus impatiens hives.

(a) During the training phase and on non-testing days, bumblebees were given access to a single foraging chamber containing a linalool-scented feeder (represented by the purple circle) for 2-3 hours per day. A 4” duct fan pulled air through the foraging chamber and vented it through a large activated-charcoal filter to prevent odor contamination of the lab space. The air-inlet to the foraging chamber was also fitted with an activated charcoal filter. Arrows indicate direction of airflow. (b) For maze trials, the single feeding chamber was replaced with a double-T maze. The maze was either unpolluted, Turf Builder-permeated, or Manzate permeated. Each chamber was treated with an appropriate amount of chemical (3.9 g Turf Builder and 0.09 g Manzate), as indicated by the orange circles. A single linalool feeder was randomly placed in a single chamber of the maze. (c). For preference experiments, the single feeding-chamber was replaced with 2 chambers in a T configuration. Each chamber contained a linalool scented feeder and a ‘pollutant’ jar. In control trials, both pollutant jars were empty. In agrochemical-trails, one randomly selected chamber’s jar contained either Turf Builder (3.9 g) or Manzate (0.09 g).