A variety of different methods for delivering noxious thermal stimuli have been developed for experimental pain studies. They generally use a contact thermode or a radiant light source to heat the skin. Progressive heating occurs till a noxious temperature is reached that provokes withdrawal response. Thus, a variety of fiber types may be stimulated before the critical population of C-fiber afferents is activated, and it is not possible to selectively stimulate different nociceptive fiber types. Laser stimulation has several advantages over contact thermodes since it can selectively stimulate either C-fibers or the more rapidly conducting A-delta fibers. The present study examines stimulus parameters to produce withdrawal using beam settings for both fiber types, and the safety factor between non-damaging noxious and damaging stimuli. A series of five ascending stimuli were delivered at 5 mm intervals along the skin on the back of anesthetized rats at or above C-fiber or A-delta stimulus intensities. Twenty-four hr later, rats were fixed with 4% paraformaldehyde, the skin containing the stimulated sites removed, paraffin embedded, sectioned and stained histochemically or using antibodies against S100A8, an abundant calcium/arachidonic acid binding protein in neutrophils and monocytes. The final stimulus in each series produced a visible blanching of the skin. This was seen as an area of dermal and epidermal protein coagulation, but without any blistering of the skin or bleeding around the edges. The damaging stimulus provoked a localized leukocyte infiltration around the edges of the coagulated region. A less abundant leukocyte infiltration could also be reproducibly seen at the next lower stimulus intensity, although no damage was appreciable either visually or microscopically. These data indicate that damage can occur at stimulus intensities approximately 10 times the withdrawal intensity and that the coagulation zone may provide a useful model for tissue injury and repair.