a very marked annual rhythm in the phenomena of seminal emissions during sleep, or, as Nelson has termed it, the ecbolic curve.  We see, also, that the great yearly organic climax of sexual effervescence corresponds with the period following harvest, which, throughout the primitive world, has been a season of sexual erethism and orgy; though those customs have died out of our waking lives, they are still imprinted on our nervous texture, and become manifest during sleep.

The fresh records that have reached me since the first edition of this book was published show well-marked annual curves, though each curve always has some slight personal peculiarities of its own.  The most interesting and significant is that of E.M. (see ante p. 116), covering four years.  It is indicated by the following monthly frequencies, summated for the four years:—­

Jan.  Feb.  Mar.  Apr.  May June July Aug.  Sept.  Oct.  Nov.  Dec.
16 13 14 22 19 19 12 12 14 14 12 24

E.M. lives in India.  April, May, and June, are hot months, but not unhealthy, and during this season, moreover, he lives in the hills, under favorable conditions, getting plenty of outdoor exercise.  July, August, and September, are nearly as hot, but much damper, and more trying; during these months, E.M. is living in the city, and his work is then, also, more exacting than at other times, September is the worst month of all; he has a short holiday at the end of it.  During December, January, and February, the climate is very fine, and E.M.’s work is easier.  It will be seen that his ecbolic curve corresponds to his circumstances and environment, although until he analyzed the record he had no idea that any such relationship existed.  Unfavorable climatic conditions and hard work, favorable conditions and lighter work, happen to coincide in his life, and the former depress the frequency of seminal emissions; the latter increase their frequency.  At the same time, the curve is not out of harmony with the northern curves.  There is what corresponds to a late spring (April) climax, and another still higher, late autumn (December) climax.  A very interesting point is the general resemblance of the ecbolic curves to the Indian conception-curves as set forth by Hill (ante p. 140).  The conception-curve is at its lowest point in September, and at its highest point in December-January, and this ecbolic curve follows it, except that both the minimum and the maximum are reached a little earlier.  When compared with the English annual ecbolic curves (W.K. and Perry-Coste), both spring and autumn maxima fall rather later, but all agree in representing the autumn rise as the chief climax.

The annual curve of A.N. (ante p. 117), who lives in Indiana, U.S.A., also covers four years.  It presents the usual spring (May-June, in this case) and autumn (September-October) climaxes.  The exact monthly results, summated for the four