officers
and men from the Euphrates to the Rhine or to the outskirts of the Sahara,
and everywhere they remained faithful to the gods of their far-away
country. The requirements of the government transferred functionaries and
their clerks, the latter frequently of servile birth, into the most distant
provinces. Finally, the ease of communication, due to the good roads,
increased the frequency and extent of travel.

Thus the exchange of products, men and ideas necessarily increased, and it
might be maintained that theocracy was a necessary consequence of the
mingling of the races, that the gods of the Orient followed the great
commercial and social currents, and that their establishment in the
Occident was a natural result of the movement that drew the excess
population of the Asiatic cities and rural districts into the less thickly
inhabited countries.

These reflections, which could be developed at some length, surely show the
way in which the Oriental religions spread. It is certain that the
merchants acted as missionaries in the seaports and places of commerce, the
soldiers on the frontiers and in the capital, the slaves in the city
homes,[4] in the rural districts and in public affairs. But while this
acquaints us with the means and the agents of the diffusion of those
religions, {25} it tells us nothing of the reasons for their adoption by
the Romans. We perceive the how, but not the why, of their sudden
expansion. Especially imperfect is our understanding of the reasons for the
difference between the Orient and the Occident pointed out above.

An example will make my meaning clear. A Celtic divinity, Epona,[5] was
held in particular honor as the protectress of horses, as we all know. The
Gallic horsemen worshiped her wherever they were cantoned; her monuments
have been found scattered from Scotland to Transylvania. And yet, although
this goddess enjoyed the same conditions as, for instance Jupiter
_Dolichenus_ whom the cohorts of Commagene introduced into Europe, it does
not appear that she ever received the homage of many strangers; it does not
appear, above all, that druidism ever assumed the shape of "mysteries of
Epona" into which Greeks and Romans would have asked to be initiated. It
was too deficient in the intrinsic strength of the Oriental religions, to
make proselytes.

Other historians and thinkers of to-day prefer to apply the laws of natural
science to religious phenomena; and the theories about the