[On cases of retinitis pigmentosa with absence of hemeralopia and with extinction of the electroretinogram. Clinico-physiological considerations on relations between adaptation to dark and the electroretinogram].