by the following contrivance: an Auer burner, provided with a sheet-iron chimney, in which a rectangular orifice, 60 mms. high and 25 mms. broad, has been cut, is enclosed in an iron lantern pierced with an opening placed in front of the chimney orifice, and stopped up by a plate of aluminium. In front of this window the small slip of paper is placed, illuminated in the manner indicated above. If, now, the rays are intercepted by interposing a sheet of lead or the hand, the small paper rectangle is seen to darken, and its contours to lose their sharpness; the light diffused by the slip of paper is thus increased by the action of "N" rays.
The following idea then presented itself: the diffusion of light is a complex phenomenon, in which the elementary fact is regular reflection, and consequently there is reason for ascertaining experimentally whether the reflection of light is, or is not, modified by the action of "N" rays. For this purpose, a polished steel knitting-needle was fixed vertically in place of the slip of paper of the former experiment; at the same time, in a box completely closed, with
