636 EEPOiiT— 1887. 



that of hydrogen ; a method which the author helieved capable of more extended 

 application in the determination of the atomic weights of other elements. Zinc was 

 prepared of ahigh degree of purity, and a given weight of the metal having been dis- 

 solved in dilute sulphuric acid, the amount of hydrogen evolved was determined by 

 volume. A solution of auric bromide or chloride as pure as possible was treated with 

 a known quantity of the same zinc, more than sufficient for the precipitation of the 

 gold; the excess of zinc was dissolved by dilute sulphuric acid, and the volume of 

 hydrogen given oft' was determined. The precipitated gold was carefully collected, 

 washed, dried, and weighed. The difference between the volume of hydrogen 

 which the zinc gave when thus partly used to replace a known quantity of gold, 

 and the volume which it would have given if replacing hydrogen only, taken in 

 connection with Hegnault's determination of the relation of weight to volume for 

 hydrogen, afforded of course the data needed for a direct comparison of the weights 

 of gold and hydrogen concerned. It was pointed out that in applying this process 

 the weight of the gold salt in solution need not be known, and that the method is 

 not dependent upon a knowledge of the atomic weight of the halogen, combined 

 with the gold, or of the atomic weight of zinc, and does not even require that the 

 zinc shall be of assured purity, provided only it be uniform in character, so that a 

 given weight of it can be depended on to yield always the same quantity of 

 hydrogen, and there be no impurities present capable of interfering with the 

 collection of the precipitated metallic gold in a state of purity. 



7. The Atomic Weight of Zirconium. By G. H. Bailey, D.Sc, Ph.D. 



The previous determinations of atomic weight of this element were made by 

 Berzelius (89-25), Hermann (88 8), Marignac (90-54). The earlier results were 

 doubtless vitiated by the presence of iron and of the cerite earths, whilst Marignac's 

 determination is open to objection from the character of the salt (potassium 

 zirconium fluoride) which he used. In the present determination, zirconia was 

 prepared from North Carolina zircons by three independent methods. This was 

 dissolved in sulphuric acid and the sulphate was crystallised out. This salt 

 becomes normal and constant in weight by heating some hours at 400°, the tem- 

 perature at which it begins to decompose being 470°. The relation of zirconium 

 sulphate to zirconia gives a ratio from which the atomic weight is calculated and 

 the value thus obtained agrees more nearly with that of Mai-ignac. The author 

 proposes to make further determinations, using the tetra bromide. 



8. Torsion Balances. By Dr. A, Springer. 



Light frames are made and then stiffened by wires or flat bands being tensioned 

 over them. The beam is firmly clamped to the bands in such a manner that its 

 centre of gravity is above its point of support; this tends to tip the beam, thus 

 equilibrating the torsional resistance of the fulcrums. We thus have the torsional 

 resistance exerted to keep the beam horizontal, and the high centre of gravity 

 tending to tip it out of the horizontal. 



The adjustment of the position of the centre of gravity is most easily made 

 by having an adjustable poise placed immediately above the central torsional 

 wire. In order to do away with the necessity of alignment of support, a secondary 

 beam is attached to the first in such a manner that both beams tending to tip in 

 the same direction remain stationary owing to their having opposite and equal 

 momenta. 



On this principle scales are constructed which can be used on rolling ships or in 

 buildings where there is considerable jarring. In all the ' Torsion Balances ' there 

 is permanence of adjustment, consequently repeated weighings will give like 

 results. 



Various ' Torsion Balances ' were shown illustrating the principles involved, 

 as well as showing how equal sensitiveness can be obtained with any load. 



