Treatment with the tracer
We fed each treated mound a nitrogen tracer mixed in sugar water. A solution with 102 mM of 15N-labeled glycine (98 atom%, Sigma-Aldrich, Inc., St. Louis, MO, USA) and 61.5 mM of unlabeled sucrose was created using distilled water. This concentration was determined based on a preliminary laboratory experiment with a small number of fire ant workers and an approximation of colony sizes in the field, which can exceed 250,000 workers within a single mound (Tschinkel, 2006). The solution was mixed in bulk at the beginning of the field experiment and frozen between uses to avoid mold growth. We filled 1-mL microcentrifuge tubes with 1 mL of the solution and stoppered each with cotton. Three of these vials were left on the surface of each treatment mound and replaced every other day for 14 days. Vials were placed directly on the mound surface to ensure that only the treated mound fed on the solution. Each treatment mound was fed a total of 160 mg of 15N-labeled glycine in 21 mL of sugar solution over a 14-day period. Fire ant workers were observed feeding on the solution, and there was evidence of mound building over the vial opening, indicating worker foraging.
We collected workers from all treated and untreated mounds once before feeding the tracer to treated mounds and once after the 14-day treatment period (Fig 1). Workers were collected by disturbing a small section of each mound and aspirating 40-50 workers for isotopic and genetic analyses (see Genetic analysis section below). Workers were frozen at -10˚C. We then transferred 10-30 workers per mound to 95% EtOH for storage prior to DNA extraction and left the remaining workers for stable isotope analysis. Workers for stable isotope analysis were never stored in EtOH to avoid possible effects of EtOH on isotopic signatures (Tillberg, McCarthy, Dolezal, & Suarez, 2006).