Specifically cytochrome C oxidase, which is part of the electron transport chain that is responsible for generation of ATP. Cytochrome C oxidase is the terminal enzyme of the electron transport chain, ultimately responsible for creating an electrochemical potential across the inner mitochondrial membrane, which drives the production of ATP. The enzyme is structurally large and complex, and possible absorbing chromophores include two heme moieties and two copper sites. Analysis of the action spectrum for cellular proliferation following laser photoirradiation and spectroscopic data on cytochrome C oxidase has suggested that the majority of photoabsorption is via oxidized CuA, reduced CuB, oxidized CuB, and reduced CuA. Our results suggest that very low densities of light may reach the brain when near infrared light is applied to the skull of stroke patients. This direct irradiation may be sufficient to change mitochondrial and neural activity. In previous studies treating strokes in rabbits and rats an 808 nm diode laser was set to give a power density of 7.5 mW/cm2 at brain level. Our measured irradiances through coronal sections of cadaver subjects are similar, albeit slightly lower. However, irradiance through frontal and temporal regions of a sagittally sectioned cadaver are approximately 10-fold lower. The average irradiance of infrared light through coronal and sagittal cadaver sections in our study is 2.43 mW/cm2. The NEST-1 trial was designed to deliver 1 Joule/cm2 to the entire surface of the cortex by treating 20 predetermined sites on the scalp for 2 minutes each. Notably, previous trials with rabbits and rats delivered a similar energy density, 0.9 Joule/cm2. Based upon our findings, in order to obtain 1 Joule/cm2 over the entire cortex surface, each site would need to be treated for an average of 6.86 minutes. More importantly, as our study illustrates, treatment times should be calculated based upon penetrance at various locations on the skull, as irradiance can vary significantly. It is important to note that the estimated energy density reaching the cortex of the patients in NEST-1, in which a coherent light source was used, was 1 J/cm2, which is much lower than the energy levels we observed using a noncoherent light source. Noncoherent versus coherent light penetrance has been the subject of some studies, and mathematical simulations, for example Monte-Carlo simulation, may provide insight into the ability of different light sources, such as laser compared to lightemitting diode, at the same wavelength to penetrate tissue. Yet, recent studies, both in vitro and in vivo, have shown that low levels of infrared light can exert effects on neural tissue. In a 2009 study, a 670 nm laser with a low peak irradiance output of 3 mW/cm2 and a low dose of 0.45 mJ/cm2 was found to stimulate nerve growth factor-induced neurite elongation in vitro, and stabilize mitochondria membrane potential in neurons exposed to H2O2. Similarly, in a 2008 in vivo study in pigmented rats, 633 nm light emitting diode treatment with power density of 2 mW/cm2 was applied via two LED arrays located 3.8 cm above the subjects’ heads for 30 minutes. The treatment increased Masitinib whole-brain cytochrome oxidase and superoxide dismutase activities in a dose-dependent manner, and prevented the decrease in visual function induced by administration of rotenone, a mitochondrial complex I inhibitor.