Needless to say the Amharic speakers also have close contact with the speakers of the other Semitic Ethiopic languages as well as of the Cushitic languages. Nearly all the Semitic Ethiopic and Cushitic languages have many Arabic loanwords for the same reasons mentioned above in connexion with Amharic. It is, therefore, often difficult to know whether an Arabic loanword in Amharic came directly from an Arabic dialect or through the intermediary of another Semitic Ethiopic or Cushitic language. Since we have no adequate means at our disposal for a historical treatment of the Semitic Ethiopic vocabulary, we shall most probably never be able to determine the precise origin of some of the words considered here as borrowings, that is to say whether they come directly from an Arabic dialect or through another Semitic Ethiopic or Cushitic language.

The sīra authorities provide a mass of contradictory data on the chronology of the Prophet's raids and the task of fitting these inconsistencies into a cogent chronological pattern is beset with difficulties. In the main sources, for example, Mūsā b. ‘Uqba, Ibn Isḥaq, and al-Wāqidī, we find noticeable divergencies in the dating of a specific event. This is so even with important incidents such as the expulsion of Banu Qaynuqā‘and a comparatively late event such as the attack on Khaybar. A textual collation of the chronological material does suggest, however, that in some cases it may be possibleto win through to a safer position.

From the studies of Zurvanism made in this century, a large measure of agreement has been reached. It is accepted that Zurvan ‘est en general le dieu du firmament lumineux et étoilé … avant tout le dieu du sort … en général regardé comme un dieu quadriforme’; and that his cult was ‘enraciné surtout dans l'lran occidental’. But a fundamental disagreement has developed over the origin of Zurvanism. H. S. Nyberg, although pointing out scrupulously that ‘le mythe zervanite n'st directement attesté que dans le mazdéisme’, and that ‘nous ne trouvons jamais un système zervanite où Ormuzd ne soit représenté’, has nevertheless maintained that Zurvan is an ancient Iranian god, whose cult, older than Zoroaster, has been partly adopted into, and partly obliterated by, orthodox Mazdeism. In this he has been followed by a number of scholars.

One of the few Old Persian words whose precise meaning has not yet been established is yakā-, the name of a material which was used in the building of Darius' palace at Susa. In the Great King's own words ‘it was brought (to Susa) from Gandara and Karmāna’.1 That a timber was meant appears from the Akkadian version (the corresponding Elamite sentence is missing), in which yakā- is translated by GIšMlŠ-MA-KAN-NA. Admittedly the Akkadian word is partly restored; V. Scheil, in his edition,2 printed it as [miš]-má-kan-na. However, both E. Herzfeld3 and F. W. Konig4 thought they could see the determinative GlŠ and a trace of MlŠ; at all events the preserved part of the word leaves scarcely any doubt as to the correctness of the restoration.

Neolithic sites of the North Karnataka region produce artifacts belonging to two distinct industries.1 The first comprises mainly blade and ‘geometric’ tools manufactured from a series of siliceous rocks. Their manufacture entails a definite tradition of core preparation and striking of bladeflakes.2 The second industry comprises core and flake tools manufactured from a group of igneous and metamorphic rocks. Their manufacture entails different techniques based upon flaking, grinding, and pecking. The artifacts to be studied here belong to the second industry.

The official chronicle of the short reign of the Emperor Shun-tsung (28 Feb. to 31 Aug., A.D. 805) was composed by Han Yü, the most famous prose writer of the T'ang dynasty. To this we owe the survival of what purports to be this chronicle, as a unique example of the ‘Veritable Records’ (shih-lu) composed in the T'ang History Office. I say purports to be because it is not at all clear what relation the version we now possess bears to that which came from Han Yüs brush. Mr. Solomon1 has given a brief, lucid analysis of the problems regarding this text but he has not, I think, examined quite all the evidence or considered all the possible interpretations of it. I shall deal more fully with this point below.

In the study of the early history of South East Asia the nature of the shipping which operated between India and the regions to the east has been the subject of much discussion. Sources for this study are rare enough and the fact that one of them, Periplus of the Erythrean Sea, § 60, is either corrupt, as Müller believed, or contains a difficult hapax legomenon, has proved something of a stumbling block. The text of the Periplus rests upon a single MS (Heidelberg: Cod. Palatinus Gr. 398 saec. X ineuntis) of which there is a fourteenth or fifteenth century copy in the British Museum (Add. Mss. 19391)

The only publications in this dialect1 of the Galla language are translations of the Gospels of St. Luke2 and St. John3 and the Acts of the Apostles.4 The orthography used in these texts, obviously, has practical aims in view and dispenses with much of the phonetic detail which would be of interest to a linguist. As far as I know, there are no published works which would give a description of this dialect. The grammars of Borello,5 Hodson and Walker,6 Moreno,7 Nordfeldt,8 and Tutschek9 all deal with other dialects of Galla, which differ from Borana in many details.