1895] MICROSCOPICAL JOURNAL. 155 



The liquid in the alum cell is about five-eighths of an inch 

 in thickness, and was prejjared b}'- mixing equal volumes of 

 cold saturated solution of alum in water and strong glycerine. 

 Very little alum separates out and any resulting color may be 

 removed by adding to the boiling hot solution enough solution 

 of sodic carbonate to produce a small amount of permanent 

 precipitate which is removed by decantation or filtering. 



This mixture seems to absorb more heat rays than does the 

 plain alum solution, so that even glycerine jelly mounts do not 

 suffer. The thickness of liquid mentioned does not suffice, 

 of course, except when the polariscope is used. The polar- 

 izer is made of twelve plates of the kind of glass used for 

 covering the best quality of lantern slides. Through the 

 kindness of Mr. Walmsley I obtained glass so colorless that no 

 tint is percej^tible in a plate five inches wide, viewed edge- 

 wise. 



If tint is unavoidable, a yellowish green is preferable ; a 

 bluish tint is the worst. 



Before assembling the plates a narrow strip of writing 

 paper was pasted around the margin on one side of each, 

 in order to prevent contact between them, and this paper, 

 as well as the general cover of paper put on to exclude 

 dust, was well wetted with a solution of corrosive sublimate 

 in alcohol. Neglect of this precaution is apt to result in 

 formation of mould between the plates. Polarization of par- 

 allel rays, incident at the proper angle, by such a pile is so 

 complete that the margin of the " dark field " is barely dis- 

 tinguishable from the unilluminated portions of the screen. 



The transmitted beam is absorbed by a dead black back- 

 ing. The parallel rays from the polarizer are converged })y 

 the second condenser-lens so as to come to a focus upon, or 

 nearly upon, the object. The lens which corrects the aber- 

 ratioa of the first condenser-lens gives the further advantage 

 of producing a blurred focus, thus obviating any distinct im- 

 age of the positive carbon ; the rays from the negative carbon 

 do not enter objectives of an inch, or less, in focal length. 



Usually an accessory (sub-stage) condenser is needed, though 

 some objectives do well without it. Simple plano-convex • 

 lenses work well. The object must be mounted on good 



