178 THE MINERAL ANALYSIS OF ROCKS. [May 1902, 
sizes of the particles, so that heavier and smaller ones might be 
mixed with lighter and larger ones. In other cases, no doubt, this 
action might have practically no effect. 
Mr. H. W. Burrows referred to the process of calculating 
chemical composition from measurement of areas of the different 
minerals seen in rock-sections, and asked the Author whether he 
considered that method approximately reliable. 
The AurHor expressed his thanks to the President for the kind 
words with which he had received this communication. It was 
gratifying to think that this method, powerful as it might already 
be, was still open to improvements; the Author hoped to continue 
his work on the subject, and meanwhile would welcome any 
criticism, especially when of the nature of experiment. There 
were necessary limitations to the application of the method, and a 
discussion of these would be found in the paper; but some which 
were suggested by Mr. Holland were he thought @ priori, and were 
disposed of in some instances by the examples that he had just 
laid before the Society. The objections alleged by the Rev. J. F. 
Blake were also theoretical ; in a well-made diffusion-column every 
mineral took up a position in strict accordance with its specific 
gravity. Mr. Burrows was to be congratulated on his successful 
employment of the well-known method of weighing areas, but the 
process was too laborious for common use. 
