Abstract This study was carried out to investigate in vivo the changes of the human lens autofluorescence (AF) with aging and cataractogenesis. Measurements were performed in the blue-green AF range (495 nm/520 nm) using a fluorometer designed, built and now clinically tested in our department. 43 random eyes of 43 healthy volunteers aged 6-86 years, five of each decade, were studied for effects of aging and 84 eyes of 84 patients with cortical, nuclear, posterior subcapsular or mixed lens opacities were studied for differences of various cataract types. The results were compared with the back light scatter values obtained by the commercially available Interzeag Lens Opacity Meter 701. Also AF and back light scatter of the lens were measured from 122 smoking males aged 57 to 76 years who participated in a cancer prevention study. The results were compared with the widely used subjective lens opacities classification system, LOCS III. In addition data was collected from 30 randomly chosen eyes of as many subjects with varying degrees of yellow-brown lens coloration in an otherwise healthy eye. We studied the influence of lens yellowing expressed by means of lens AF on visibility of retinal nerve fiber layer in black-and-white images. Lens AF profile consists of anterior and posterior peaks and a central plateau. The height of the anterior peak was used as a measure of the maximum AF value. The square root of the ratio between the posterior and the anterior AF peaks was used for estimating the lens transmission. Our technique was highly reproducible. The coefficient of variation was 3.9% for maximum AF and 2.9% for the lens transmission index. Both the maximum AF and light scatter were exponentially increased with age (r = 0.95 and 0.94, respectively; p < 0.0001). According to the regression line of AF begins to increase in early childhood. It appears by extrapolation to be absent at birth. In contrast light scatter in the lens was present even in young children. The lens transmission for blue-green light, determined from the lens AF curve, was almost unchanging with age up to 60 years. Thereafter it decreased rapidly and the interindividual variation increased.In cataractous lenses the mean AF and scatter values differed statistically significantly from those of age matched healthy controls. The highest AF values were measured in nuclear cataracts where AF was also related to visual acuity and an increasing yellow-brown colour of the nucleus. About half of the total variation of the transmission index values could be accounted for by changes in nuclear colour as assessed by the LOCS III grading system. The transmission index provided a more precise prediction about nuclear colour and opalescence than age or light scatter did. In cortical cataracts the AF curve was low and flattened and the maximum AF value was significantly lower than in the age matched control eyes. The highest light scatter values were measured from cortical cataracts, but the correlation between LOCS III cortical grades and light scatter values was rather weak. Posterior subcapsular cataracts cannot be quantified either with AF or with light scatter measurements. Lens yellowing, expressed as lens AF, had an actual effect on retinal nerve fiber layer visibility. AF measurements provided a better prediction about the visibility score than age or visual acuity did. The results of the present study indicate that the lens autofluorescence measurement may be a useful additional tool together with a subjective grading system in the follow-up of optical changes occurring in the nuclear region of the lens.