Such pairs of months exist where greater seasons are involved in the determining of the moons, and they are in fact convenient, since their use obviates the unfortunate circumstance which has been a source of great confusion to primitive peoples, viz. that a natural phase from which it is the custom to name a month may fall on the border-line between two moons. So long as the description of the months remains quite fluctuating and occasional, this and similar inconveniences do not make themselves felt, but a very natural development leads to a conventionalising of the series of months. In common speech a selection among the various names of months unconsciously takes place, so that those prevail which relate to more important occupations and natural phases. Thus arises a fixed, or tolerably well fixed, series of months, such as appears in most of the reports handed down to us. CHAPTER VIII. OLD SEMITIC MONTHS. 1. BABYLONIA. In the much disputed questions of the ancient Babylonian astronomy and calendar the non-expert is in a situation of despair: for whoever cannot himself make use of the sources is referred to the often directly contradictory statements of the experts. I cannot however shirk the task of investigating whether in Babylonian calendric systems traces of the primitive time-reckoning are not also to be found. Unfortunately I cannot limit myself to matters upon which a certain unity of opinion prevails, but must also touch upon burning questions, such as the intercalation. What is here offered is in the nature of things only an attempt: but I may perhaps be allowed to express the hope that competent specialists, not led astray by chronological hypotheses, may afterwards observe how far the few but obvious characteristics of the primitive time-reckoning recur also in the Babylonian system.