254 REPORTS ON INVESTIGATIONS AND PROJECTS. 



trace of water could be expelled from this salt at 300 — the highest tem- 

 perature to which it could be subjected without decomposition. The uncer- 

 tainty, small in itself, is nevertheless one of great importance in this connec- 

 tion. Preliminary experiments made with the help of T. S. Woodward, and 

 already referred to in the last report, seemed to indicate the purity of the 

 salt, but these were not definitive. Accordingly, with the help of Marshall W. 

 Cox, the author instituted new and more critical experiments. It was found 

 that between 300 and 400 , at which temperature the lithium perchlorate 

 suffers slight loss of oxygen, no water was given off by the salt; and also 

 that when the substance was still further heated until almost wholly decom- 

 posed, no water could be found in the gas evolved. This discovery indicates 

 that the salt as formed in the earlier investigation had been as pure as it 

 could possibly be made, and affords substantial support for the atomic 

 weight of silver, 107.871 (0= 16.000), obtained in that investigation. The 

 matter will be prosecuted further in the future. 



(2) Atomic weight of uranium: 



With the assistance of Dr. R. Adams the investigation of the atomic 

 weight of uranium was begun. This atomic weight, the largest of all, has 

 acquired peculiar interest in the last few years because of the suspicion that 

 the element may spontaneously decompose into helium and radium. If this 

 is the case, its atomic weight should be the sum of that of radium and some 

 multiple of that of helium. Hence, an exhaustive revision of the earlier 

 work of the author in conjunction with Dr. B. S. Merigold seemed desirable, 

 because that work was only preliminary in its nature. The problem is a 

 difficult one and no final results are as yet to be recorded, but progress has 

 been made in discovering the errors and side-reactions which must be taken 

 into account. 



(3) Atomic weight of aluminum: 



The investigation mentioned in the preceding paragraph led to the devis- 

 ing of apparatus which could be employed profitably in other researches of 

 a similar kind. One of the most pressing cases to which the apparatus is 

 applicable is that of aluminum, the atomic weight of which has not been 

 seriously studied for over 30 years. Hence, with the help of C. C. Wallace, 

 the investigation of the atomic weight of aluminum was begun. Good 

 progress has been made, not only in preparing pure aluminum salts, but also 

 in effecting preliminary trials of the details of experimentation. This inves- 

 tigation, like the preceding, will be continued in the immediate future. 



(4) Atomic weight of carbon, as found from molecular weights of carbonates: 

 Often in the past the carbonates of the more electropositive metals have 



been used as starting-points for the determination of atomic weights, but 

 usually the results have been of doubtful value because of uncertainty as to 

 the stability and purity of the substances in question. New knowledge of 



