148 Prof. J. Joly on a Method of Investigating the 



probably, be made good in the calibration constant as deter- 

 mined with the same reagent. Its use would possess the 

 advantage that boiling over could not occur. On the other 

 hand, the melt is not so readily removed from the boat or 

 crucible ; solution appearing to be necessary. 



The foregoing results appear to indicate that this method 

 of collecting the emanation during decomposition and fusion 

 is more searching than the method by solution. Its effec- 

 tiveness is doubtless clue to the small bulk of the materials 

 dealt with, their very high temperature, and, in the case of 

 the use of carbonates, to the intense effervescence attending; 

 decomposition. Acid rocks yield their emanation somewhat 

 more freely than do basic rocks. This maybe due to the fact 

 that in the case of basic melts there is often a gravitational 

 separation in the crucible ; a heavy layer, possessing very 

 great viscosity even at the highest available temperatures,, 

 forming at the bottom. The use of boracic acid prevents this 

 segregation^ and gives a homogeneous melt as well as intense 

 effervescence. 



While there appears no doubt as to the more searching 

 nature of the new method and little difficulty in accounting 

 for its effectiveness, it is not easy to explain why the solution 

 method gives a relatively deficient yield. One would fancy 

 that the mode of determining the calibration constant would 

 compensate for effects tending to conceal the emanation ; the 

 added radium being exposed to the same causes of deficiency 

 as affect the original radium in the rock. This may not be 

 true, however. There might be a more or less constant 

 amount of emanation held back in rock solutions, and this 

 loss would then fall upon the yield when the rock content 

 only was being investigated. The calibration experiment 

 made with the chemically rich rock solution, in this case, 

 would only reveal this phenomenon in so far as it might 

 increase with the quantity of radium present. An actual 

 concealment of some 30 per cent, on the yield of emanation 

 from pure aqueous solutions is, in fact, observed. 



There is another possibility. There may be more ema- 

 nation in the alkaline solution than is boiled off. This would, 

 a jiriori, seem a likely explanation ; but I have, so far, failed 

 to find any support for it in experiments in which the alkaline 

 solution was tested after acidification. But again, there may 

 be causes leading to the failure of this test. 



The fusion method seems not only more searching, it is less- 

 open to risk of contamination, for the preparation is limited 

 to the pulverization of the rock. Its rapidity leaves nothing 

 to be desired. If the crucible or boat is chilled after an 



