MICROCHEMISTRY. I47 



is to the organic laboratory what spectrum analysis is 

 to the inorganic." When it is necessary to separate 

 two closely related isomers, to determine the identity 

 of two substances prepared by diiJerent methods, to 

 distinguish differences of elementary composition from 

 those due to allotropism, or to detect minute admixtures 

 of an impurity, it frequently occurs that purely chemical 

 methods are tedious and complex, while microscopical 

 examination furnishes a rapid and easy solution. In 

 general the method of study consists in allowing the 

 substance in question to crystallize in contact with a body 

 of known composition, with which it is to be compared. 

 This may be done in three ways — by adding to one 

 substance, in a melted condition or in solution, a crystal 

 of the other, by preparing a mixture of both in a liquid 

 condition and allowing the mixture to cool, and best> 

 perhaps, by crystallizing from films of the two sub- 

 stances brought just in contact with each other. 



"When solid substances are to be compared, a small 

 portion of one is placed under a cover-slip and heated 

 till it melts, the amount of material being so small as 

 not to reach the edge of the cover. A grain of the second 

 substance is then placed just outside the cover-slip, and 

 in turn melted so as to flow under and join the first along 

 a line which can be observed under the microscope. (In- 

 cidentally it may be noted that by heating two bodies 

 simultaneously under the microscope a good measure is 

 obtained of slight differences in melting-points. Thus 

 the similar crystals of dinitrobenzol and dinitrotoluol 

 may be easily distinguished.) As two substances cool 

 and recrystallize, the phenomena along the contact zone 



