traight instead of being
slightly broken, then the curve would, in its chief features, be
geometrically symmetrical; and this symmetry appears to me to afford a
convincing proof of the representative accuracy of the curve. We see that
the month is divided into five periods; that the maxima occur on the
following pairs of days: the 19th-20th, 13th-14th, 25th-26th, 1st-2d,
7th-8th; and that the minima occur at the beginning, end, and exact middle
of the month. There have been many idle superstitions as to the influence
of the moon upon the earth and its inhabitants, and some beliefs
that--once deemed equally idle--have now been re-instated in the regard of
science; but it would certainly seem to be a very fascinating and very
curious fact if the influence of the moon upon men should be such as to
regulate the spontaneous discharges of their sexual system. Certainly the
lovers of all ages would then have "builded better than they knew," when
they reared altars of devotional verse to that chaste goddess Artemis.

THE WEEKLY RHYTHM.

We now come to the third branch of our inquiry, and have to ask whether
there be any weekly rhythm of the sexual activity. _A priori_ it might be
answered that to expect any such weekly rhythm were absurd, seeing that
our week--unlike the lunar month of the year--is a purely artificial and
conventional period; while, on the other hand, it might be retorted that
the existence of an _induced_ weekly periodicity is quite conceivable,
such periodicity being induced by the habitual difference between our
occupation, or mode of life, on one or two days of the week and that on
the remaining days. In such an inquiry, however, _a priori_ argument is
futile, as the question can be answered only by an induction from
observations, and the curves on Chart 11 (_A_ and _B_) prove conclusively
that there is a notable weekly rhythm. The existence of this weekly rhythm
being granted, it would naturally be assumed that either the maximum or
the minimum would regularly occur on Saturday or Sunday; but an
examination of the curves discloses the unexpected result that the day of
maximum discharge varies from year to year. Thus it is[383]

Sunday in 1888, 1892, 1896.
Tuesday in 1894.
Thursday in 1886, 1897.
Friday in 1887.
Saturday in 1893 and 1895.

Since, in Chart 11, the curves are drawn from Sunday to Sunday, it is
obvious that the real symmetry of the curve is brought out in t