12 ; REPORT—1845. 
Another question, which can only be decided by experiment, is whether a trans- 
parent dielectric in a highly polarized state affects light transmitted in the same 
manner as a uniaxal crystal*. All analogy would certainly lead us at least to look 
for such an action in a plate of glass of which the particles are kept ina constrained 
state by means of opposite electrical charges on the two faces, especially when we 
consider that the constraint may be elevated to such an extent as to make the sub- 
stance be on the point of cracking. 
Before concluding this abstract, it may perhaps be permitted to call attention to 
another object of experimental research. All the measurements of Coulomb have 
been made solely for the purpose of comparing electrical forces with one another, 
but to complete the theory, we should have the means of comparing electrical forces 
with weights, to which every other kind of force is ultimately referred. For this 
purpose a standard intensity must be chosen, and the diminution of atmospheric pres- 
sure at a point of conducting surface, possessing this intensity, either determined 
by direct measurement, or deduced from experiments in which the repulsion between 
bodies charged to a given intensity is measured by weights. 
. The standard intensity is furnished by the result of Mr. Harris mentioned above, 
and might be taken as the intensity immediately before a spark, in a given state of | 
the atmosphere. The series of experiments necessary to complete the investigation 
would be of an extremely delicate nature, and might be long and laborious; but if 
the result were arrived at, and if the laws of action of dielectrics were thoroughly 
known, the experimental elements of the theory would be complete. 
Remarks on the Periodicity of Magnetic Disturbances. 
By the Rev. H. Luoyn, F.R.S., MRA. 
When we examine, for the first time, the chart of the changes of one of the 
magnetic elements during a day of disturbance, we do not hesitate to pronounce 
that the causes which produce these changes, so apparently capricious, belong to 
the class which, from our ignorance of their laws, we are accustomed to denominate 
“accidental ”’ or “‘ irregular.’”’ Experience, however, has shown that these pheno- 
mena, and therefore also the forces which produce them, are subject to laws, which 
require multiplied observations alone for their development. A few months of 
systematic observation is sufficient to show that these apparently abnormal move- 
ments of the magnet recur more frequently at certain hours of the day than at others. 
Prof. Kreil seems to have been the first to notify this remarkable fact. In a letter 
addressed to M. Kupffer, dated in January 1839, he observes, that ‘‘all hours of 
the day do not appear to be equally favourable to the development of this phzno- 
menon ;”’ that disturbances begin “much more frequently in the evening than in 
the morning hours,”’ and “ hardly ever begin in the latter hours of the forenoon.” 
In a letter addressed to Col. Sabine, dated in July 1840, Prof. Kreil has entered 
more minutely into the question, with the light of the observations of an additional 
year. He there observes, that “the least disturbance takes place in the declination 
from 8 to 10 a.m., and the greatest from 8 to 10 p.m.;”’ that “the declination is 
increased by the disturbances of the forenoon and middle of the day, and diminished 
by those occurring in the evening hours ;” that the effect of disturbances upon the 
horizontal intensity is, in general, a diminution of that element, this diminution 
being however more considerable ‘‘ during the hours of the night and morning, than 
in the forenoon and afternoon.” 
A more elaborate examination of this question has been since made by Col. Sabine, 
in the discussion of the results of the first two years’ observations, made at the Mag- 
netic Observatory of Toronto, under the direction of Lieut. Riddell. The mode of 
examination is, for the most part, the same as that of Prof. Kreil, namely, to sepa- 
rate the individual results, which differ from the monthly mean, corresponding to the 
same hour, by a quantity exceeding a certain arbitrary limit; to treat them as the 
effects of perturbing causes ; and to examine the frequency of their occurrence at the 
* Since this paper was read the author has found that Mr. Faraday had previously pro- 
posed and examined experimentally the question here suggested, arriving only at negative 
results. See ‘Experimental Researches,’ § 955, 
