28 PROCEEDINGS OP THE AMERICAN ACADEMY 



T'° + 273 = ^,{T°-\- 273) W+iN-N')w ^^ _j_ j- yo _ ^.o-j ^^ 



With this formula, it is now easy to compute the values for each divi- 

 sion of our arbitrary scale. "We cannot, however, depend absolutely 

 on the result, as there are several causes which will differ with each 

 instrument, and of which we can take no account. It is therefore best 

 to observe, with the instrument, two or three well-established boiling 

 points, which will give us fixed points by which we can correct the 

 table, and we shall then have an instrument whose precision is fully 

 equal to that of a mercury thermometer. 



It is, of course, very desirable that the temperature of reference T° 

 should be always the same and as invariable as possible. This is best 

 accomplished, as above suggested, by maintaining a circulation of water 

 through a glass jacket surrounding the stem of the instrument and 

 enclosing also a small mercury thermometer, which is best tied to the 

 stem. By selecting the temperature of reference a little higher than 

 the highest temperature of the water supplied by the laboratory hy- 

 drants, it is easy to maintain the required temperature within a degree 

 by regulating the flow. The instrument may then be adjusted to the 

 tubulature of a retort and used in fractional distillations. 



VI. Revision op the Atomic Weight of Cadmium. 



By Oliver W. Huntington, of the Senior Class. 



The method adopted by Professor Cooke for verifying the value of 

 the atomic weight of antimony, described in this volume (page 16), 

 proved to be so definite and conclusive, that it seemed desirable to 

 apply it in all other cases to which it was suited, in order not only to 

 verify the received values of the atomic weights, but also to test more 

 fully the hypothesis of Prout, an hypothesis to which recent investi- 

 gation and speculations have given renewed interest.* 



The method of Professor Cooke is applicable to all those elements 

 of which a pure, stable, well-defined, and soluble bromide can be pre- 



* See Revision of the Atomic Weight of Aluminum, by J. W. Mallet, Philo- 

 sophical Transactions, part iii. 1880 : also various papers by J. Norman Lock» 

 yer on the disassociation of the chemical elements in "Nature" and in the 

 Proceedings of Royal Society ; also Cooke's Chemical Philosophy, revised edi- 

 tion, page 272. 



