404 
PROFESSOR J. W. MALLET ON THE ATOMIC WEIGHT OF GOLD. 
attempting any nuraei'ical estimate of relative value, by so-called “weighting” of the 
results in calculation. 
3. In connection with each process there should be careful study of the reactions 
depended upon for the final determination of an atomic weight, looking especially to 
the possibility of the occurrence of secondary or subsidiary reactions. 
4. Each process adopted should be as simple as possible, both in the nature of the 
chemical reaction or reactions and in the known liability to merely manipulative 
errors. 
5. Each process should be carried out with, in some experiments larger, and in 
others smaller, quantities of material. But, on the whole, the quantities used should 
be kept within such limits as are most likely to admit of most accurate determinations 
being had under the conditions of the special process. 
6 . In the reactions depended upon only such other elements should be concerned as 
maybe counted among those of which the atomic weights are already known with the 
nearest approach to exactness. 
7. It is particularly desirable that, if possible, the atomic weight to be investigated 
shall be, by at least one process, compared directly with that of hydrogen, now almost 
universally taken as the basis for the whole list of the elements. It is remarkable for 
how very few of the elementary substances—not more than three or four—this direct 
comparison has heen accurately made. 
8 . In the greater number of the processes available for atomic weight determina¬ 
tions the comparison with hydrogen must perforce be made indirectly. When this is 
the case, it is desirable that as few other elements as possible, the assumed atomic 
weights of which will have to be taken into account, shall be involved in each single 
reaction depended upon. 
9. In selecting different processes to be applied to the determination of the atomic 
weight of a given element, in order that the results may check each othey it is desir¬ 
able that, not the same, but as many different other elements as possible, shall be 
concerned in the several reactions, provided all such elements count amongst those of 
which the atomic weights may be considered in the first rank as to the accuracy with 
which they are known. 
Means and Aletliods of Weighing Einployedi 
These were in the main the same as those which I had in former years used in 
determining the atomic weight of aluminum. 
The balance chiefly used, made by Becker, was carefully cleaned, and all its parts 
adjusted, especially as to the position of the centre of gravity for each load to be 
used. A second balance by the same maker, of larger size, capable of taking a load 
of a kilogramme in each pan, was employed in weighing certain of the solutions 
experimented on, and was in like manner carefully adjusted and tested. All 
