Three monkeys were treated with krypton red laser photocoagulation in a manner simulating treatment of a choroidal neovascular membrane. One eye of each monkey was treated by employing immediately the power level needed to produce a lesion judged biomicroscopically to be of sufficient intensity (Method A). The fellow eye was treated by starting with lower power levels and then gradually increasing to the level needed to produce a similar-appearing lesion (Method B). Parameters of spot size and treatment duration were kept constant. Animals were sacrificed three days, eight days, and four months after treatment. Both methods of treatments spared the overlying ganglion cell layer in all fundus locations, but Method A induced cell loss in the inner nuclear layer while Method B spared it. Since reaching the clinical endpoint for Method B required fewer spots at the highest power level, the number of spots at the highest power level is probably the most important factor in determining degree of inner retinal absorption. Since both methods of treatment resulted in satisfactory closure in the choriocapillaries, the histopathologic differences may not be clinically important.