Mr. H. F. Talbot's Experiments on Light. 327 



ment. A drop of saturated solution of sulphate of copper was 

 precipitated by ammonia. The precipitate was again dissolved 

 by more of the alkali, taking care to add no more than was 

 requisite for that purpose. Some strong aceticacid was then 

 added, and the whole stirred together with a glass rod, upon 

 a plate of glass, which was then laid upon the stage of the mi- 

 croscope. In about half a minute spontaneous crystallization 

 took place of a very beautiful salt, which I presume to be the 

 acetate of ammonia and copper, in rhomboids of a deep blue 

 colour, and very perfect in shape, insomuch that among 

 many hundreds of them the microscope detected scarcely any 

 that had any flaw or defect of transparency. In shape these 

 crystals are not unlike those of common sulphate of copper, 

 but they are distinguished from them by a decisive character, 

 namely, their deep blue colour, even when of the very smallest 

 dimensions: for the crystals of sulphate of copper, when of 

 this degree of minuteness, are perfectly white. Now, when 

 the incident light is polarized, it causes a singular change in 

 these crystals ; for one half of them become green, the other 

 half acquiring a purer blue than before. But on turning round 

 the plate of glass that supports them, the blue ones are seen 

 to change their tint to green, while the green in their turn 

 become blue: thus exhibiting the dichroism of which Sir David 

 Brewster was the first discoverer; and exhibiting it, let it be 

 observed, in crystals too small to be seen with the naked eye, 

 and without the observer's having the trouble of cutting and 

 polishing their surfaces. 



Potash may be substituted for ammonia in this experiment, 

 but soda will not answer the purpose. 



§ 2. On Photometry. 



The subjects of optical inquiry are so diversified and di- 

 stinct that I have found it convenient to arrange the experi- 

 ments I am relating under several separate heads, which have 

 perhaps, little or no connexion with each other. Those men- 

 tioned in the last section were made lately, while, on the con- 

 trary, those in the present section, relating to photometry, were 

 made nine years ago, and should have been published at that 

 time, except that I wished to render them more worthy of 

 publication by following up the many other trains of experi- 

 ment which they suggested. An article in a foreign journal 

 having recalled my attention to this subject, I will proceed to 

 give a short account of the view I took of it formerly, and 

 which I have seen no reason to alter since. 



Photometry, or the measurement of the intensity of light, 

 has been supposed to be liable to peculiar uncertainty. At 



