Fletcher — A Refined Method of obtaining Sublimates. 465 



Comparative Standards. 



It is possible in this method to keep more oi- less permanent records of 

 the tests which may serve as comparative standards. The initial brightness 

 of most of the deposits is transitory, the colour fading more rapidly in moist 

 atmospheres. 



Sublimates, however, may be liermetically sealed by surrounding with a 

 ridge of seccotine, upon which a plate of glass is pressed in presence of 

 nitrogen or carbon dioxide. 



Limits of Seitsibility. 



The delicacy of this method in detecting the presence of very minute 

 quantities of volatile impurities in solution, and the limits to which the 

 detection could be thus carried under the most favourable circumstances, 

 are evident in the following experiments. The results show the limiting 

 amounts of known substances in solution which can be detected. It is 

 possible that, with substances yielding the more intense colours with the 

 iodine-plate, it will be possible to identify very small quantities of substances 

 in solution. 



A small quantity of solution of the substance is transferred to the 

 platinum ribbon support from a capillary pipette. The ribbon is heated 

 gently to evaporate the water, when the substance can often be seen as a 

 dull stain on the bright surface of the platinum. A glass cover-plate is 

 polished and placed over and very close to the ribbon, which is rapidly 

 heated to whiteness, and allowed to cool. When the sublimate is very 

 thin indeed, it is at this stage nearly invisible, and may be demonstrated 

 either by moistening with the breath or polishing with a soft cloth, wlien the 

 deposit which is invisible by transmitted light will become visible by reflected 

 light against a dark background. 



By this method it is possible to detect the presence in a drop of solution 

 of 4-5 X 10"' gram of lead nitrate, or therefore of 2'8 x 10"' gram of 

 metallic lead, and assuming all the lead to have been deposited as white 

 oxide, the sublimate would contain about 3 x 10"' gram of lead oxide. The 

 deposit was not uniform in thickness, but thinned off indefinitely at the 

 edges, occupying an area of about 7 sq. mm., and the average thickness 

 therefore could not have exceeded 4'6 X 10"' cm. — a thickness of less than 

 one-tenth that of the thinnest gold-leaf. 



In a similar manner it is possible to distinguish easily the sublimate 

 obtained from a small drop (00009 c.c.) of a solution of arsenic in dilute 



