Objective Studies of primate vision often use either luminance or chromatic modulation in an attempt to isolate postreceptoral mechanisms.There is substantial evidence that luminance or chromatic modulation, either red-green (long (L) and middle (M) wavelength-cone opponent) or blue-yellow (short (S) wavelength-cone), was processed in separable psychophysical channels, and the physioIogical substrates of these channels lie in the magnocellular (MC), parvocellular (PC) and koniocellular (KC) pathways respectively.But the role of these systems in spatial vision is still controversial.Methods We have recorded responses of cells in V1 of the awake macaque monkey to a novel grating stimulus drifting across their receptive fields which we call compound grating.It consists of equiluminant red and green bars separated by dark areas.The luminance component of this compound grating is twice the temporal (and spatial) frequency of the chromatic component.Results The ratios of the second harmonic (luminance) to the first harmonic (chromatic) response is bimodally distributed, with ratios less than one for PC cells and greater than one for MC cells.In contrast, the distribution of ratios for cortical cells is continuous.By comparing responses to conventional luminance, chromatic gratings and compound gratings, the input to many cortical cells could be identified as predominantly MC (62% of cells), a small group as predominantly PC (21% of cells), with some cells receiving combined input (17% of cells).Conclusion These results confirm previous reports demonstrating a surprising amplification of the MC pathway at the cortical level (60% of cortical cells) and a significant group of luminance/chromatic cells.