THE LINEAR FORCE OF GROWING CRYSTALS 315 



ment in a question of fact to remain unverified where a principle of 

 such far-reaching importance to geologists was at stake. The issue 

 could have been put to the test by anyone in a few hours without 

 special facilities of any kind. 



When, however, a distinguished physical chemist (Boeke, in his 

 admirable book, Grundlage der physikalisch-chemischen Petrographie, 

 1915, p. 328) accepts their conclusion without test, and casts out 

 the "linear force of growing crystals" from geological calculation 

 entirely, a form of protest appears necessary. This outcome is the 

 more unfortunate because it must be patent to a physical chemist 

 that Bruhns and Mecklenburg did not in fact repeat our simple 

 experiment at all, but substituted one which is not, by itself, con- 

 clusive upon the point at issue; and, further, because all of their 

 recorded experimental evidence is entirely in accord with our 

 experience and conclusions. There appears to be no contradiction 

 of fact, but only one of appropriate interpretation; the conditions 

 which govern the behavior of a single loaded crystal are modified 

 when an unloaded crystal is introduced into the same solution, as 

 was done by Bruhns and Mecklenburg. 



Now upon the main question of fact as to whether a crystal of 

 alum or other substance will or will not lift a load when immersed 

 in its saturated solution in an open vessel, nothing is simpler than 

 to repeat the experiment which we described. An ordinary open 

 crystallizing dish, an alum crystal placed on the bottom of it and 

 covered with a saturated solution of the same liquid, and an ordi- 

 nary brass weight of one or two hundred grams upon the crystal, 

 together with a simple apparatus to be found in any laboratory for 

 measuring the thickness of the crystal before and after the experi- 

 ment, provide all the equipment necessary to establish or disestab- 

 lish the fact of growth in the direction of the load. Two fairly 

 typical cases follow: 



These experiments are so straightforward, and withal so conclu- 

 sive in their results, that it would hardly seem possible to go astray; 

 nevertheless, Bruhns and Mecklenburg, in the paper above referred 

 to, have denied their validity. A series of measurements taken 

 from the paper of Bruhns and Mecklenburg is quoted in Table II. 1 



1 Op. cit., p. 100. 



