Mr. J. F. Daniell on Crystallization. 43 



cannot fail of proving fertile of important consequences to the 

 corpuscular philosophy. 



There is a great and natural reluctance in the mind to give 

 up an hypothesis ingeniously constructed, like that of M. 

 Haiiy, so as not only to present in a general point of view a 

 great number of particular facts, but to enable us to reason 

 from the known to the unknown, and actually to predict facts 

 before trial ; and it is from this cause, I think, that the sphe- 

 roidal view of crystallization has not hitherto received all that 

 consideration which it appears to me to deserve. 



It is, however, by no means uncommon in physics to find 

 two theories maintained as to the origin of natural phenomena, 

 both of which cannot of course be the real laws of nature, but 

 each of which will enable us to generalize and group the facts 

 together with accuracy. I need only refer to the two theories 

 of electrical phenomena and the two theories of heat. 



When two hypotheses thus run parallel to each other, and 

 each explains a great many facts in common with the other, 

 such unforeseen evidence in favour of one, as I have been 

 endeavouring to explain, is of the utmost consequence. But 

 it is of scarcely less importance to show, if possible, the con- 

 nexion of one with the other, which enables them to apply to 

 so many circumstances in common. 



Now it appears to me that this is not difficult with regard to 

 the two theories of crystallization. 



Those who have attentively studied the great work of M. 

 Haiiy must be well aware that he could not have been insen- 

 sible to the difficulties which attended his system, when viewed 

 as an account of the real process of nature ; but, whilst in 

 doubt with regard to this ultimate object of inquiry, he most 

 philosophically adopted a temporary substitute for truth, which 

 was capable of leading him by legitimate reasonings to conclu- 

 sions in exact accordance with observations so numerous as 

 fairly to embrace the whole range of phenomena which the 

 theory was intended to account for. 



The fact is, that, so far from insisting on the real existence 

 of his primitive forms and integrant molecules, as many of his 

 followers injudiciously have done, he very early in his reason- 

 ing points out the difficulty of the ambiguous choice, and of 

 the vacant interstices j and shows that the whole purpose of 



