ly. This was a precaution
that occurred to me, as there was a possibility that the vertical
diameter of the retina might be more or less sensitive than the
horizontal one."

Astronomers will recognize in this the first suggestion of the processes
which have led to important results in the hands of Dr. OTTO STRUVE and
others in the comparison of the measures of double stars by different
observers, each of whom has a personal habit of observation, which, if
not corrected, may affect his results in the way which HERSCHEL was
striving to avoid.

_Researches on the Relative Brightness of the
Stars: Variable Stars._

No research of HERSCHEL'S was more laborious than the elaborate
classification of the stars according to their comparative brightness,
which he executed during the years 1796 to 1799. It was directly in the
line of his main work--to find out the construction of the heavens.

His first paper had been upon the variable star _Mira Ceti_. Here was a
sun, shining by its native brightness, which waxed and waned like the
moon itself. This star is periodic. It is for a long period invisible to
the unassisted eye. Then it can just be seen, and increases in
brightness for a little over a month, and attains a maximum brilliancy.
From this it decreases for nearly three months, and after becoming
invisible, remains so for five or six months. Its whole period is about
333 days. Are all other stars constant in brightness? The example of
_Mira Ceti_ and of other known variables makes this at least doubtful.
But the sun itself may vary for all that we know. It is a simple star
like the rest.

This question of variability in general is an important one, then.
It can only be tested by making accurate catalogues of the relative
brilliance of stars at various times, and by comparing these. No such
general catalogue existed before HERSCHEL'S time, and led by the
discrepancies in isolated cases, which he found between his own
estimates and those of his predecessors, he made from observation a
series of four catalogues, in which were set down the order of sequence
of the stars of each constellation.

The method adopted by HERSCHEL was perfectly simple in principle, though
most laborious in practice. Suppose any number of stars, A, B, C, D, E,
. . . etc., near enough to each other to be well compared. The process
consists simply in writing down the names of the stars, A, B, C, etc.,
i