TRANSACTIONS OF SECTION B. 457 
platinum black and with palladium black are employed. Palladium coated 
with palladium black gives in both 0.1nH.SO, and 0.1nHCl a value 4 to 
5 millevolts too high; which is permanent even when the hydrogen is 
allowed to pass over the electrode for hours. But gold and platinum 
coated with either platinum or palladium black give in a very few 
minutes in both HCl and H,SQ,l, values concordant among themselves to 
probably less than 0°05 of a millevolt. A piece of gold or platinum wire is 
as good as a large piece of foil. It has usually been considered that the 
hydrogen gas electrode in 0.1nHCl is not so constant as in 0.1nH,SO,. But 
this does not seem to be the case; at all events if precautions against the 
possible presence of arsenic be taken. It seems probable from Bredig’s work 
that the smallest trace of arsenic would ‘ poison’ the platinum or palladium 
black, and according to Crookes’ it is practically impossible to purchase 
HCl free from arsenic. When this is borne in mind it seems preferable to 
use 0.1nHCl rather than H.SO,, because the dissociation constant is so much 
better known and the ions produced are so much simpler. It is not possible 
to get a good coating of platinum black on a very pure specimen of platinum 
foil if the platinum chloride solution employed be also very pure. A trace 
of lead acetate is usually added to the solution therefore, and it is necessary 
to remove every trace of this from the electrode after coating, or good values 
will be unobtainable. Heating in a solution of oxalic acid and then in 
nitric acid removes all lead quite satisfactorily. ; 
3. The Measurement of Rotatory Dispersion. 
By T. Martin Lowry, D.Sc. 
Measurements have been made with light of the following twenty-six 
wave-lengths : 
Li 6708 Na 5893 Tl 5351 (flame spectra). 
Hg 5790 5769 5461 4359 (enclosed arc). 
Cd 6438 5086 4800 4678 Zn 6364 4811 4722 4680. 
Cu 5782 5700 5218 5153 5106 4705 4651 4587 4378. 
_ A photographic method has also been devised by means of which measuze- 
ments may be made throughout the transmitted spectrum, so far as this can 
be recorded on a photographic plate. 
From the above list the seven wave-lengths shown in heavy type have 
been selected for general use, the green mercury line, Hg 5461, being selected 
on account of its brilliance and purity as principal standard in place of 
the sodium doublet, 5890 5896. 
Measurements have been made of the optical and magnetic rotations 
produced by quartz and by a series of optically active alcohols, acids, and 
esters. In the case of quartz there is an absolute agreement between the 
two dispersions, but every optically active liquid that has been examined 
shows a divergence between the two series of values, the optical dispersion 
being usually, but not always, higher than the magnetic dispersion. It 
is possible that the identity of the optical and magnetic dispersions in 
crystals is an indication that the magnetic rotatory power of liquids depends 
upon a spiral packing of the molecules of the same general character as 
that which produces the optical rotatory power of quartz, but the evidence 
at present available is not sufficient to justify a definite pronouncement. 
4. Mercurous Sulphate for Standard Cells. 
By Cuartss J. J. Fox, B.Sc., Ph.D. 
It is now well established that the chief cause of inconstancy in cadmium 
cel's is usually due to irregularities in the mercurous sulphate. F, E. Smith 
+ Select Methods of Analysis, 1905, p. 564, 
