Abstract
Having accurate estimates of the absorption spectra of pre-receptoral filters in the eye, and the lens in particular, is required for relating visual sensitivity of the eye measured with psychophysics to the absorption spectra of the visual pigments. Spectral density curves of the lens have relied on transmission measurements of material excised from cadavers; however, these have the disadvantage of the possibility of errors introduced by post mortem changes. In particular, with increasing time after death excised lenses become progressively cloudier. Ideally, the lens should absorb short wavelengths and introduce minimal scatter. Single pigment absorbers follow Boltzman statistics, which are modeled well by a single Gaussian. However, the human lens measurements from cadaver tissue exhibit a non-Boltzman statistics curve between the main pigment cut-off and the transition into the longer wavelengths. The possibility that there are errors in the human lens density curves in common use due to time-dependent post mortem changes have not been well investigated. We measured spectral transmission of lenses from three species of genus Macaca within a maximum 25 minutes post euthanasia (n=30). We also derived an in vivo estimate of the human lens using data of scotopic sensitivity from Crawford (1949, Proc. Phys. Soc. B, 62:321; n=50, age<30) and a theoretical rhodopsin photopigment template. Both results from the macaque and the human in vivo sensitivity show that the shape of the lens is consistent with a single pigment fit by a Gaussian. Time post-euthanasia induces Rayleigh scatter elements even before the cataract is visible.