We made a comprehensive study, involving observations on 45 marmosets, of the effects on ocular growth and refraction of wearing spectacles from the ages of 4-8 weeks. This period was within the period early in life when the eye grows rapidly and refraction changes from hyperopia to its adult value of modest myopia. In one series of experiments we studied the effect of lenses of powers -8, -4, +4 and +8D fitted monocularly. In another series of experiments we studied the effect of lenses of equal and opposite powers fitted binocularly, with the two eyes alternately occluded, so as to give an incentive to use both eyes, and in particular to accommodate, for at least part of each day, through the negative lens. The vitreous chamber of eyes that wore negative lenses of -4D or -8D, combined with alternate occlusion, elongated more rapidly than that of the fellow eye (negative lens eye-positive lens eye, 0.21 +/- 0.03 mm (S.E.M.), P < 0.01 and 0.25 +/- 0.06 mm, P < 0.05, respectively) and became relatively more myopic (2.8 +/- 0.26D, P < 0.01 and 2.4 +/- 0.61D, P < 0.05 respectively). Eyes that wore -4D lenses monocularly elongated more rapidly and became myopic than fellow eyes. Eyes that wore +4D or +8D lenses were less strongly affected: animals that wore +8D lenses monocularly (without alternate occlusion) developed a slight relative hyperopia (0.99 +/- 0.21D, P < 0.01), with the more hyperopic eyes also slightly shorter (0.09 +/- 0.05 mm) than their fellow eyes, but eyes wearing +4D lenses were not significantly different from their fellow eyes. Animals that wore -8D lenses monocularly (without alternate occlusion) developed a slight relative hyperopia after three weeks of lens-wear (0.85 +/- 0.26D, P < 0.05). These were the only eyes that responded in a non-compensatory direction to the optical challenge of spectacle wear, and we interpret this effect as one due to visual deprivation. After the removal of lenses, the degree of anisometropia slowly diminished in those groups of animals in which it had been induced, but in the three groups in which the largest effects had been produced by lens-wear the overall mean anisometropia (0.68 +/- 0.24D, P < 0.01) and vitreous chamber depth (VCD) discrepancy (0.09 +/- 0.03 mm, P < 0.01) were still significant at the end of the experiments, when the animals were 273 days old. The reduction of anisometropia in these groups was associated with an increase in the rate of elongation of the vitreous chamber in the eyes that had previously grown normally i.e. the less myopic eyes grew more rapidly than their fellow eyes: in the seven weeks following lens-wear these eyes became more myopic and longer than normal eyes (refraction P < 0.001; VCD P < 0.001). Control experiments showed that occlusion of one eye for 50% of the day had no effect on eye growth and refraction, and therefore that alternate occlusion itself had no effect.