Many comparative neurobiological studies seek to connect sensory or behavioral attributes across taxa with differences in their brain composition. Such studies can only be interpreted in a meaningful way if the general brain-body relationships are known on a larger taxonomic scale. Recent studies in vertebrates suggest cell number and density may be better correlated with behavioral ability than brain mass or volume, but few estimates of such figures exist for insects. Here we use the isotropic fractionator method to estimate total brain cell numbers for 29 genera of Hymenoptera spanning seven subfamilies. We find estimates from using this method are comparable to traditional, whole-brain cell counts of two species and to published estimates from established stereological methods.  We present allometric scaling relationships between body and brain mass, brain mass and nuclei number, and body mass and cell density and find that ants stand out from bees and wasps as having particularly small brains by measures of mass and cell number. We find that Hymenoptera follow the general trend of smaller animals having proportionally larger brains. Smaller Hymenoptera also feature higher brain cell densities than the larger ones, as is the case in most vertebrates, but in contrast with primates, in which neuron density remains rather constant across changes in brain mass. Overall, our findings establish the isotropic fractionator as a useful method for comparative studies of brain size evolution in insects. 

Following measurements of head width and body mass, heads were severed from bodies and brains dissected out in wax or on SYLGARD® coated petri dishes (Sigma-Aldrich, St. Louis, MO, USA). Brains, including the subesophageal ganglion (SOG) and retinas, were removed as carefully as possible to prevent breakage or loss of tissue. Broken samples or those clearly missing optic lobes, antennal lobes, or the SOG were discarded. 

Brains stored in cacodylate buffer were rinsed three times for 10 minutes each in PBS, then transferred to a 0.5 mL tissue grinder tube (Cole-Parmer, Vernon hills, IL, U.S.A.). The volume of dissociation solution (40 mM sodium citrate in 1% Triton-PBS), dilution used, and time brains were homogenized depended on the size of the brain (electronic supplementary table 2). After homogenization the sample was transferred to a 3.7 mL glass vial or 1500 μL centrifuge tube. The homogenizer was rinsed three times with a small amount of PBS and this volume added to the total dilution. Following homogenization and dilution, 50 μL of 4’,6-diamidino -2-phenylindole (DAPI; Sigma-Aldrich, D9542), in PBS (1:100,000) was added for every 0.5 mg of brain up to 200 μL (table S2). The sample was wrapped in aluminum foil and placed on the rotator for 30 minutes and, immediately before cell counting, was vortexed (Vortex Genie 2; VWR, Radnor, PA, USA) to distribute nuclei homogenously. 

Following labeling with DAPI and vortexing,10 μL subsamples were loaded into a hemocytometer (Neubauer Improved Bright-Line Cell Counting Chamber; Jiangsu Co. Ltd, China) and counted. Cells were counted under epifluorescence using a 40x objective on an Axio Imager upright microscope (Zeiss, Göttingen, Germany). The hemocytometer has two grids that each contain four sets of sixteen 0.25 mm x 0.25 mm squares at a depth of 0.1 mm, hence each square represents a volume of 0.00625 mL. Nuclei were counted in each grid and the number used to approximate the total number of nuclei in the brain considering the total volume comprising the homogenized brain. Sixteen small squares were counted per grid and the mean represents one subsample. Ten subsamples (grids) were counted per brain.