TRANSACTIONS OF THE SECTIONS. 7 



some of its leading properties with regard to liglit, and he found, 1st, 

 that the light reflected from a plate of mica so prepared, (which light 

 is very intense,) is but feebly polarized in the plane of incidence ; and 

 2nd, that the reflection so far resembles that at metallic surfaces, that 

 when plane-polarized light is reflected from it, the plane of reflection 

 being inclined to that of primitive polarization, the light is found to be 

 elliptically polarized. The latter fact he stated to be in a great measure 

 explained theoretically by a remark made by Professor Lloyd, to whom 

 he had mentioned the experiment. The observation, and also the 

 theory of it, cannot fail, he thinks, to be important in illustrating the 

 nature of metallic reflection, which is at present so actively discussed. 



On the bending of silvered Plate Glass into Mirrors. 

 By James Nasmyth. 



The author described a simple, and, so far as he is aware, original 

 mode of forming concave mirrors of vast size for reflecting telescopes 

 from disks of silvered plate glass, which by means of the pressure of 

 the atmosphere, he bends at pleasure in a cold state to any required 

 degree of curvature (within reasonable limits). Mr. Nasmyth brought 

 to the meeting a disk of silvered plate glass, 3 ft. 3 in. diameter, 

 which he could at pleasure bend and unbend into a concave mirror, by 

 simply withdrawing some of the air from behind its surface ; the disk 

 of plate glass being cemented round the edge on a circular plate of 

 cast iron, so that the air-tight cavity thus formed behind the outward 

 surface of the glass permits the pressure of the atmosphere to act and 

 press it into a concave mirror, the instant any portion of the air behind 

 is withdrawn. Any required degree of curvature can be retained by 

 simply preventing the return of the air behind by means of a stop- 

 cock. 



On Photometry, or a mode of measuring diffuse daylight comparatively, 

 at any time and place. By Dr. Andrew Ure. 



When lights of different intensities have the same quality of tint, or 

 tone of colour, they may be measured relatively to each other, by the 

 relative depths of shadow which they project upon a white wall or 

 screen, from an opaque body interposed. When the tint of the light, 

 however, is very different, as with the bluish flame of gas, the vivid 

 white of an argand lamp fed with oxygen (called the Bude light), or 

 the gray light of the sky, it becomes very difficult to measure the re- 

 spective intensities of such lights, by comparing the shadows which 

 they project, with the shadows projected from the flame of a standard 

 wax caudle, or a mechanical lamp. Dr. Ure experienced this diffi- 

 culty, of late, upon two interesting occasions. The first was in trying 

 to measure the relative illuminating powers of the Bude light, in sub- 

 serviency to my examination before the Select Committee of the House 

 of Commons on lighting the House. The second was in estimating 



