The application of fundus reflectometry to the study of the choroidal circulation is based upon the principle of projecting a light of known intensity into the eye through the pupil and measuring the amount of light which returns through the pupil from the fundus oculi. It is assumed that the light passes through the retina and choroid, is reflected by the sclera and traverses the retina and choroid again before leaving the eye. If this were so, fundus reflectometry could be used at wavelengths which would give information about the amount of blood in the choroid, the degree of oxygenation of the blood, and the rate at which colorimetric changes occur in the fundus.Various types of apparatus have been designed to examine the choroidal circulation by fundus reflectometry, some being modifications of the apparatus used previously to study the bleaching of visual pigments. The continuous recording of changes in the amount of light reflected from the fundus requires steadiness of the eye and of the head, otherwise large irregularities will be introduced into the recordings. In attempts to overcome this problem, devices have been produced which can be mounted on the cornea.The result obtained from the human eye during alteration of the oxygen content of the circulating blood (by hyperventilation and apnoea, or by a very brief period of nitrogen-breathing, or during oxygen-breathing) was of particular interest in assessing the validity of fundus reflectometry in quantitative studies of the choroidal circulation. When the oxygen content of blood was reduced there was a rise in optical density in the red region of the spectrum and a fall in the blue region; the corresponding fall in reflection of red light from the fundus of the human eye could be elicited consistently by fundus reflectometry but the expected rise in reflection of blue light was absent. With the same apparatus and the same human subject, changes in the reflection of blue light by the fundus, induced by bleaching of visual pigments, could be repeatedly recorded. The most likely explanation of this finding is that light of longer wavelength penetrates the choroid and is reflected back from the sclera whereas light of short wavelength is reflected from an interface somewhere between the layer of rods and cones and the blood vessels of the choroid. The differences in pathways taken in the fundus by lights of various wavelengths introduces serious complications into fundus reflectometry as a densitometric procedure, but despite difficulties in its quantitative application the technique is capable of yielding limited information about the choroidal circulation, particularly in the albino rabbit eye. It would appear that observations of the rate of changes in the pigmented human eye are valid, especially those that compare the timing of events in the two eyes of an individual, as, for example, following the intravenous injection of Evans blue or indocyanine green, and it is in the area of investigation of carotid stenosis and similar disorders that the techniqu...