aracter which long clung to them; and the improved
definition of their outlines has not, so far, brought them into
disaccord with truth. The most promising hypothesis of the kind is due
to Professor Bigelow of Washington. His able discussion of the eclipse
photographs of January 1, 1889,[589] showed a striking agreement between
the observed coronal forms and the calculated effects of a repulsive
influence obeying the laws of electric potential, also postulated by
Huggins in 1885.[590] Finely subdivided matter, expelled from the sun
along lines of force emanating from the neighbourhood of his poles, thus
tends to accumulate at "equipotential surfaces." In deference, however,
to a doubt more strongly felt then than now, whether the presence of
free electricity is compatible with the solar temperature, he avoided
any express assertion that the coronal structure is an electrical
phenomenon, merely pointing out that, if it were, its details would be
just what they are.

Later, in 1892, Pupin in America,[591] and Ebert in Germany,[592]
imitated the coronal streamers by means of electrical discharges in low
vacua between small conducting bodies and strips of tinfoil placed on
the outside of the containing glass receptacles. Finally, a critical
experiment made by Ebert in 1895 served, as Bigelow justly said, "to
clear up the entire subject, and put the theory on a working basis."
Having obtained coronoidal effects in the manner described, he proceeded
to subject them to the action of a strong magnetic field, with the
result of marshalling the scattered rays into a methodical and highly
suggestive array. They followed the direction of the magnetic lines of
force, and, forsaking the polar collar of the magnetised sphere,
surrounded it like a ruffle. The obvious analogy with the aurora polaris
and the solar corona was insisted upon by Ebert himself, and has been
further developed by Bigelow.[593] According to a recent modification of
his hypothesis, the latter appendage is controlled by two opposing
systems of forces; the magnetic causing the rays to diverge from the
poles towards the equator, and the electrostatic urging their spread,
through the mutual repulsion of the particles accumulated in the
"wings," from the equator towards either pole. The cyclical change in
the corona, he adds, is probably due to a variation in the balance of
power thus established, the magnetic polar influence dominating at
minima, the electrostatic at