The U.S. Forest Service has a long history of providing termiticide efficacy data used for product registration and labeling. Four primary test sites (Arizona and Florida, Mississippi, and South Carolina [hereafter southeast]) have been used for this purpose. Various parameters of termite attack at water-only control plots were examined in this study to assess the relative pressures of termites at each site. Termiticide studies installed between 1971 and 2001 by using ground board (GB) and concrete slab (CS) test methods were included. GB control plots were attacked 85% of the time in the southeast, about twice the rate observed in Arizona (43%). CS plots were attacked 59–70% of the time in the southeast, significantly higher than in Arizona (43%). Termites were slower to initiate attack at control plots in Arizona compared with the southeast, and they were up to twice as slow at GB controls. Once initial attack began, GB plots were reattacked at higher percentages in the southeast (89–90%) than in Arizona (67%). Reattack at CS plots ranged from 65% in Arizona and South Carolina to 76% in Mississippi. Termites caused less damage to wooden blocks in control plots in Arizona than the southeast. Attack rates at controls generally declined during the 1990s, but these rates have rebounded since 2000, except at CS plots in Arizona and South Carolina. Statistical analysis of attacks at plots treated with chlorpyrifos, cypermethrin, fenvalerate, and permethrin also was undertaken. Time to initial termite attack (failure) of the organophosphate chlorpyrifos was generally shorter in Arizona than in the southeast, whereas time to initial attack in plots treated with one of three pyrethroids (cypermethrin, fenvalerate, and permethrin) was generally longer in Arizona.