As regards lunar periodicity, E.M., has summated his results in a curve, after the same manner as Mr. Perry-Coste, beginning with the new moon.  The periods covered include 54 lunar months, and the total number of discharges is 176; the average frequency is about 3 per month of twenty-eight days.  The curve, for the most part, zigzags between a frequency of 4 and 9, but on the twenty-fourth day it falls to 1, and then rises uninterruptedly to a height of 11 on the twenty-seventh day, falling to 2 on the next day.  Whether a really menstrual rhythm is thus indicated I do not undertake to decide, but I am inclined to agree with E.M. himself that there is no definite evidence of it.  “It looks to me,” he writes, “as if the only real rhythm (putting aside the annual cycle) will be found to be the average period between the ecboles, varying in different persons, but in my case, about nine and one-eighth days.  May not the ecbolic period in men be compared to the menstrual period in women, and be an example of the greater katabolic activity of men?  There is the period of tumescence, and the ecbole constituting the detumescence.  The week-end holiday would hasten the detumescence, but about every third week-end there would tend to be delay to enable the system to get back into its regulation nine or ten days’ stride.  This might possibly be the explanation of the curves.  The recent emissions were nearly all involuntary during sleep.  Age may have something to do with the change in character.”

E.M.’s curves frequently show the influence of weekly periodicity, in the tendency to ecbole on Sunday, or sometimes on Saturday or Monday.  In recent years there has been some tendency for this climax to be thrown towards the middle of the week, but, on the whole, Wednesday is the point of lowest frequency.

In another case, the subject, A.N., who has spent nearly all his life in the State of Indiana, has kept a record of sexual manifestations between the ages of 30 and 34.  The data, which cover four years, have not been sent to me in a form which enables the possibility of a monthly curve to be estimated, but A.N., who has himself arranged the data on a lunar monthly basis, considers that a monthly curve is thus revealed.  “My memoranda,” he writes, “show that discharges occur most frequently on the first, second, and third days after new moon.  There is also another period on the fourteenth and fifteenth, which might indicate a semi-lunar rhythm.  The days of minimum discharge are the seventh, eighth, twenty-second, and twenty-third.”  It may be added that the yearly average of ecbolic manifestations, varying between 50 and 55, comes out as 52, or exactly one per week.

A weekly periodicity is very definitely shown by A.N.’s data. 
Sunday once more stands at the head of the week as regards
frequency, in this case very decisively.  The figures are as follows:—­