SEPTEMBER 24, 1897. ] 
the ohm was nearly one part in 2,000 smaller 
than that usually accepted, but the uncer- 
tainty of the measurements was due rather 
to discrepancies in the comparisons of the 
standard coils than to want of consistency 
in the readings given by the instrument. 
Professor Ayrton also communicated a 
paper by Mrs. Ayrton on the ‘ Relations be- 
tween Arc Curves and Crater Ratios with 
Cored Positive Carbons,’ which is being pub- 
lished elsewhere. Mrs. Ayrton is well- 
known as an authority on these subjects, 
but the paper is one which cannot be ade- 
quately explained without the curves by 
which it was illustrated. Papers were also 
contributed by Cerew and Basquin, ‘On the 
Are;’ by Willard E. Case, on ‘Gas Bat- 
teries,’ and Professor J. C. MacGregor, on 
the ‘State of Ionization of Solutions.’ 
In the department of general physics, 
there were a number of interesting papers 
on acoustics and on astronomical and mag- 
netic observations, but the writer was unable 
to obtain any adequate abstracts or reports. 
On Wednesday, Professor Ramsay read a 
paper on the refractivity of mixtures of 
gases. According to his observations, the 
refractivity of a mixture could not be de- 
duced by the usual mixture formula. The 
discrepancy between theory and observation 
in the case of a mixture of hydrogen and 
helium amounted in some proportions to as 
much as 3%, with oxygen and nitrogen to 
0.35%. These differences were systematic, 
and appeared to be too large to be due to 
observational errors. They indicate that 
the refractivity is not accurately propor- 
tional to the partial pressure, but that there 
is a deviation analogous to the deviations 
from Boyle’s law, and apparently in a sim- 
ilar direction. It must be remembered, 
however, that the refractivity of helium is 
so phenomenally low as to make these 
measurements extremely difficult. 
Professor O. Lodge introduced a discus- 
sion on ‘ Zeeman’s Discovery of the Effects 
SCIENCE. 
471 
of Magnetism on Spectral Lines.’ These 
effects have been recently described in a 
communication by Professor O. Lodge to 
the Royal Society of London, but it would 
appear from the discussion which took place 
in Section A, that the effects are so small 
even in the strongest fields, and are so diffi- 
cult of study, that the physical explanation 
of the effect is still a matter of speculation. 
Professor Ayrton communicated three 
papers by himself and J. Mather, on the 
use and sensibility of galvanometers, a sub- 
ject upon which they are universally recog- 
nized as leading authorities. He explained 
the principle of the constant total current 
shunt, as applied to ballistic galvanometers, 
proposed a standard method of measuring 
the sensibility, and discussed the question 
of short versus long-period galvanometers 
for very sensitive zero-tests. He also ex- 
hibited a very simple and sensitive galva- 
nometer of the D’Arsonval type, with a coil 
resistance of only 1.9 ohms, time of com- 
plete oscillation 7.5 secs., which gave a 
deflection of 36 millimetres at 2 metres for 
one microvolt. 
A paper on the comparison of certain 
Thermocouples with the platinum resist- 
ance pyrometer was read by Mr. H. M. 
Tory, of McGill College. He showed that 
the parabolic formula could be regarded 
only as a rough approximation to the law 
of variation of the E.M.F. of a thermo- 
couple. Using a platinum-rhodium couple 
enclosed in the same tube as the platinum 
pyrometer, he obtained very consistent re- 
sults up to 1,000° C., showing deviations 
from the parabola of the same type as with 
the copper-iron couple at low temperatures. 
He showed that the assumption of a straight- 
line formula for the platinum-rhodium 
couple, which is often taken as sufficiently 
correct, may lead to errors of the order of 
100° C. 
Professor Callendar read a paper by him- 
self and H. T. Barnes, of McGill College, on 
