By the year 1873 all scholars were agreed that the cuneiform script was not invented by the Semitic Babylonians, but by a people who spoke an agglutinative as opposed to an inflexional language, a language which was therefore, at least in this respect, akin to the Tartar languages. In the following year however Joseph Halévy, the famous French Semitist, started a theory which denied the existence of a Sumerian language altogether, and explained the ideographic texts in the bilingual inscriptions already alluded to, as a secret writing intelligible only to the priests; but primâ facie the theory lacked probability and even plausibility. Halévy, it is true, propounded his theory at a time when the study of Sumerian was in its infancy, though it can hardly be said to have grown out of its childhood even at the present day, but this notwithstanding, it would be indeed singular if the priests took the precaution to enshrine their secret lore in cryptic language, and then frustrated themselves by subscribing an Assyrian translation. Moreover many of the Sumerian inscriptions treat of such very ordinary matters, that it is extremely difficult to see how it could have been necessary to employ a cryptic language to conceal them. A more ready explanation is to be found in the theory accepted by the majority of scholars to-day,—that the Sumerian language existed side by side with Semitic Babylonian, and was used much as Latin is to-day. One class of tablet especially easily distinguishable by its shape and size is that comprising legal contracts for the exchange of land, cattle and property of every description. They are small in size, oblong in shape, both sides being slightly concave, and the whole not unlike a small narrow pillow in general appearance. Many of these contract tablets were enclosed in clay envelopes to ensure their preservation. When a contract was effected by the Babylonians, the contracting parties had recourse to a legal or priestly official, and the terms of the agreement were set forth on a clay tablet which was deposited either in the temple or the record chamber: it was furthermore protected by a clay envelope upon which the terms inscribed on the contract tablet were copied in duplicate; thus every precaution was taken to secure the preservation of the original document. Sometimes the text on the envelope varies somewhat from that contained in the document itself, and in such cases the envelopes therefore have more than a purely archaic interest, and are of actual linguistic value. One or two copies were made of the contract and were kept by either or both of the contracting parties. The deed was subscribed by the witnesses, one of whom was the scribe who drew up the document and sealed it. The seal was generally affixed by rolling a small cylinder seal over the tablet while still moist, though sometimes a three-sided clay cone received the impress of the seal, and this cone was attached to the tablet by means of a reed