38 REPOKT— 1873. 



fine horizontal slit, the observer of course looking tlirough the slit ; and in some 

 of the observations this card was fixed in an adjustable stand, and the slit brought 

 into coincidence with the real image before looking through. The red and blue 

 were not much altered by the introduction of the card ; but yellow could with diffi- 

 culty be obtained, the yellow previously obtained having in fact been a highly com- 

 posite colour. 



The apparatus was left to itself for several days ; and its focal length was found 

 to be continually increasing ; that is to say, the real image receded further and 

 further from the vessel, the average recess (estimated very roughly) being about a 

 foot per day, till it reached the wall, which was 10 feet distant. 



The experiment was repeated, first with solution of sugar of lead, and secondly 

 with solution of alum, in place of solution of salt ; but the original experiment gave 

 the finest displays of colour. 



There is no difficulty in explaining the phenomena above described. They are 

 mainly due to the bending of rays towards that side on which the index of refrac- 

 tion is greatest (which in the above instances is the lower side), and to the fact 

 that this bending is greatest for the rays of shortest wave-length. The magnitude, 

 however, of the chromatic effiBct is very startling ; and I am not aware that any 

 such residts have been previously recorded. 



Possibly the increase of focal length in such an arrangement as is above described 

 may be found to furnish a convenient test of the rapidity of liquid diflusion. 



On Irradiation. By Professor G. Foebes. 



Photographs of Fluorescent Substances. Exhibited by Dr. Gladstone, F.R.S. 



These photographs were of the same nature as those exhibited at the Meeting in 

 1859, to show that the alteration of the refrangibility of the extreme rays of the 

 spectrum by fluorescent substances reduces their chemical activity. But as it had 

 been objected that the lessened photograpliic eflect might be due to a change of 

 surface through wetting the paper and coating it with a salt, a crucial experiment 

 was made by writing on a piece of white paper with black ink, bisulphate of qui- 

 nine, bisulphate of potash, common salt, and pure water. When this was photo- 

 graphed, the writing in water or in the non-fluorescent salts was not perceptible, 

 but the fluorescent quinine was strongly rendered, though not so strongly as the ink. 

 In another photograph, however, two glasses filled respectively with ink and with 

 a very sti-ong but colourless solution of quinine, came out equally almost black. 



On the Dresser-Rutherford Diffraction-grating. 

 By J. NOEMAN LOCKTEE, F.R.S. 



On the Relation of Geometrical Optics to other Branches of Mathematics and 

 Physics. By Professor Cleek Maxwell. 



The author said that the elementary part of optics was often set before the student 

 in a form which was at once repulsive to the mathematician, immeaning to tlie phy- 

 sical inquirer, and useless to the practical optician. The mathematician looked for 

 precision, and found approximation ; the physicist expected unity in the science, 

 and foimd a great gulf between geometrical and physical optics ; and the optician 

 found that if he had to design a microscope, he was expected to combine the ana- 

 lytical power of a Gauss with the computative skill of a Glaisher before he could 

 make head or tail of the formulae. The author maintained that elementary optics 

 might be made attractive to the mathematician by showing that the correlation 

 between the object and the image is not only an example, but the fundamental tj'pe 

 of that principle of duality which was the leading idea of modern geometry. The 

 object and image were homographic figures, such that every straight line or ray in 

 t-he one was represented by a straight line or ray in the other. The relations between 



