64 SHCTIONAL ADDRESSES. 
clange. This argument is sound, but confirmatory evidence by quanti- 
tative analysis would have rendered the matter certain. 
The other question under dispute, viz.: whether actual fusion had 
taken place or only re-crystallisation in a pasty mass, was of a higher 
order of difficulty, and neither in Hall’s time nor for a century later 
were means available definitely to settle it. 
During the whole of the nineteenth century this controversy lasted 
and no satisfactory conclusions were reached. Those who tried to 
‘repeat Hall’s experiments with porcelain tubes, gun-barrels, and iron 
cylinders met the same difficulties as he did, and weze on the whole less 
successful in overcoming them. In most cases their gun-barrels burst 
or the method of stopping them failed. It became clear that a coherent 
mass could be obtained from powdered chalk or pure carbonate of lime 
at a red heat without great pressure, but no one obtained really con- 
vincing evidence of fusion. The problem remained practically as Hall 
had left it. 
The twentieth century, however, has witnessed a tremendous im- 
provement in our methods of tackling such questions as these, and the 
result has been that a new department of physico-chemical or experi- 
mental petrology has been opened up and already possesses a large and 
most interesting literature. It is really the old experimental geology 
of Sir James Hall, developed almost beyond recognition. Essentially 
three factors have produced this result. One is the application of the 
electric furnace, so powerful and at the same time so compact and easily 
managed. By its means temperatures from 1000° to 1600°C. are 
easily. obtained, and as many silicates and other minerals have fusion 
points between those limits their behaviour in the molten state and 
during crystallisation and cooling becomes accurately observable. The 
second factor is the invention of the electric pyrometer, by which tem- 
peratures up to the melting-point of platinum can be observed instan- 
taneously and continuously with an accuracy of one or two degrees 
centigrade. The third important element which has determined the 
recent progress of knowledge in this field is the theoretical mathematical 
researches of such men as Willard Gibbs, Roozeboom, Schreinemakers, 
and Smits, which are so full and clear that in many respects they are 
far in advance of the experimental results. To these we may add the 
continual improvement in microscopic methods of determining minerals, 
and the advance in knowledge of crystallography, optics, and analytical 
chemistry. 
Among workers in this field it is generally agreed that only the purest. 
chemicals or minerals should be used, as the presence even of traces 
of impurity may greatly modify the phenomena, and the interpretation 
of the results is so difficult that unnecessary complications must be 
studiously avoided. 
The behaviour of CaCO, under heat and pressure is really a question 
of two components, CaO and CO,, one of these being solid and very 
infusible, the other a gas at ordinary temperatures. We may simplify 
it by regarding the system for our present purposes as consisting of 
CaO and CaCO, with CO, as a volatile constituent, arising from the 
dissociation of CaCO, at certain temperatures and pressures 
