_Instinct
Sexuel_, pp. 209, 221, and his "Priapisme Epileptique," _La
Medecine Moderne_, February 4, 1899.) The epileptic convulsion in
some cases involves the sexual mechanism, and it is noteworthy
that epilepsy tends to appear at puberty. In modern times even so
great a physician as Boerhaave said that coitus is a "true
epilepsy," and more recently Roubaud, Hammond, and Kowalevsky
have emphasized the resemblance between coitus and epilepsy,
though without identifying the two states. Some authorities have
considered that coitus is a cause of epilepsy, but this is denied
by Christian, Struempell, and Loewenfeld. (Loewenfeld, _Sexualleben
und Nervenleiden_, 1899, p. 68.) Fere has recorded the case of a
youth in whom the adoption of the practice of masturbation,
several times a day, was followed by epileptic attacks which
ceased when masturbation was abandoned. (Fere, _Comptes-rendus de
la Socitete de Biologie_, April 3, 1897.)

It seems unprofitable at present to attempt any more fundamental analysis
of the sexual impulse. Beaunis, in the work already quoted, vaguely
suggests that we ought possibly to connect the sexual excitation which
leads the male to seek the female with chemical action, either exercised
directly on the protoplasm of the organism or indirectly by the
intermediary of the nervous system, and especially by smell in the higher
animals. Clevenger, Spitzka, Kiernan, and others have also regarded the
sexual impulse as protoplasmic hunger, tracing it back to the presexual
times when one protozoal form absorbed another. In the same way Joanny
Roux, insisting that the sexual need is a need of the whole organism, and
that "we love with the whole of our body," compares the sexual instinct to
hunger, and distinguishes between "sexual hunger" affecting the whole
system and "sexual appetite" as a more localized desire; he concludes that
the sexual need is an aspect of the nutritive need.[59] Useful as these
views are as a protest against too crude and narrow a conception of the
part played by the sexual impulse, they carry us into a speculative region
where proof is difficult.

We are now, however, at all events, in a better position to define the
contents of the sexual impulse. We see that there are certainly, as Moll
has indicated, two constituents in that impulse; but, instead of being
unrelated, or only distantly related, we see that they are really