ever, being common to all parts of
Babylonia, we can hardly suppose that any temple should have existed
which did not have its exorcising formulas. In the combination of these
formulas into a ritual, due consideration would naturally be had to the
special gods invoked, the obvious result of which would be to produce
the long lists of deities that are often embodied in a single
incantation. The details of this process can of course no longer be
discerned, but the inevitable tendency would be towards increasing
complications. The effort would be made to collect everything, and from
all known quarters. Hence the heterogeneous elements to be detected in
the texts, and which, while adding to their interest, also increase the
difficulty of their interpretation. In consequence of the presence of
such heterogeneous elements, it is difficult to determine within an
incantation series any guiding principles that prompted the collectors.
Still we can often distinguish large groups in a series that belong
together. So we have whole series of addresses to the fire-god ending
with incantations, and again a series of descriptions of the group of
seven spirits serving a similar purpose as introductions to
incantations, but we cannot see on what grounds the transition from one
subject to the other takes place. Indeed the transitions are generally
marked by their abruptness.

The only legitimate inference is that the main purpose of the collectors
of incantation texts was to exhaust the subject so far as lay in their
power. They included in their codes as much as possible. The exorciser
would have no difficulty in threading his way through the complicated
mass. He would select the division appropriate to the case before him
without much concern of what preceded or followed in the text. Moreover,
these divisions in the texts were clearly marked by dividing lines,
still to be seen on the clay tablets. These divisions correspond so
completely to divisions in the subject-matter that the purely practical
purpose they served can hardly be called into question, while at the
same time they furnish additional proof for the compiled character of
the texts.

As for the date of the composition of the texts, the union of the
Babylonian states under Hammurabi, with its necessary result, the
supremacy of Marduk, that finds its reflection in the texts, furnishes
us with a terminus _a quo_ beyond which we need not proceed for _final_
editing. On the oth