D. H. Müller says of the Śheri poems which he published in 1907 that they are ‘in einem älteren Mehri-Dialekt abgefasst, der dem Überlieferer nicht ganz verstandlich war’. He goes on to say that despite his best efforts he has not been able to make complete sense of all of the poems he wrote down.

The Amharic verb expresses various aspects of action by a system of derived stems. The basic stem is modified either internally (by various infixes) or externally (by the addition of one of a number of prefixes). To these derived forms there are usually applied terms taken directly from, or modelled on. traditional European grammatical terminology. These terms, such as passive, reflexive, causative, factitive, etc., do not fully describe the function of the particular forms to which they are applied. They can also lead to an incorrect understanding of the exact function of a particular form. However, they have been and continue to be used if only through long custom and familiarity. In the Amharic verbal system there are two excellent examples of this, namely those two forms called by some scholars causative and factitive and by others first and second causative, respectively; they are, of course, those derived forms with prefix /a-/ and /as-/ respectively. Whilst in many examples these preformatives can be found to express what causative and factitive (or any other such name) may be defined as meaning, there are in these two instances just as many examples to the contrary. For example, ‘he fed’, does express a causative turn of the base form ‘he ate’, but ‘he acquired’ and the derived ‘he studied’ do not appear to have such a clear relationship.

The Library of the Wellcome Institute is devoted to the history of medicine and related sciences. Apart from some 250,000 printed books, it contains about 10,000 manuscripts, half of which are in Oriental languages.

In 1970 I was entrusted with describing the Ethiopian manuscripts belonging to this library. There were 34 of them: 17 mostly of religious content and completely unrelated to the interests of the Institute, 16 magical scrolls, and one manuscript mainly of divinatory content. Without constituting medical manuscripts in the strict sense of the term, the scrolls are designed to combat demons and diseases and to keep people in good health. They can be called magico-medical scrolls and belong to the field of ethnomedicine. The scrolls are written in Gə'az with strong Amharic influence. At the suggestion of Dr. F. N. L. Poynter, Director of the Wellcome Institute, the Wellcome Trust decided to donate the first group of manuscripts to the British Museum, to whom 14 were presented in June 1970 and 3 others in June 1971. All these manuscripts have been described by the present writer in the Catalogue of the Ethiopian manuscripts in the British Museum acquired since 1877 (in preparation). The present catalogue notes contain only the description of the 16 scrolls and the divinatory manuscript.

The four preceding articles in this series have been devoted to the chapter ‘Verbs’ of Das chwaresmische Sprachmaterial einer Handschrift der ‘Muqaddimat al-adab’ von Zamaxšari, I, attention being focused chiefly on the doubtful or erroneous transliterations of the Xw. glosses provided by the editor, Johannes Benzing. In this final article we return to the incomplete chapter on ‘Nouns’ which begins the book. Many of the glosses in Muq., 1–72 (Benzing, 1–60) remain sadly obscure, being insufficiently pointed hapax legomena of the type CCByk, BC'Dyk. Even some of these entries show the inadequacy of Benzing's transliteration scheme, e.g. TBBc 22.1 to indicate ‘BBBc with the points of t, probably for the first letter’,RNB 25.4 for ‘RBB with one point, probably for -n-’, B'BBwk 40.3 which could well be B'mwk. kCYNyk 26.5 but BCBByk 44.4 for the identical sequence of undotted letters, šY'MY'Nyk 49.1 for what could as well be šΛS'Byk, NTY 58.4 ‘owl’ for what is undoubtedly, if surprisingly, nty. In every such case reference to the facsimile is absolutely essential, and occasionally profitable, as will be seen below.

Burrow and Emeneau (1961, p. xiii, n. 3) have observed that ‘Ta. and Ma. seem to have evidence for two phonemes [/n/, the dental and /ṉ/, the alveolar]a in PDr., but the matter is not yet clear’. This is mainly due to the non-availability of data about the distribution of these phonemes in those two languages and the allophonic distribution, if any, in other languages. With the help of increased material available now, an attempt is made in this paper to find out the history of the development of the two nasal phonemes in Ta. and Ma. and to establish their allophonic nature in PDr. Since definite information about the phonetic nature of the phoneme /n/ is not available for languages such as Kui, Parji, Nk. [Ch.] etc., this study is incomplete to some extent.

The graph in Classical Chinese represents three different words, a noun (fu1/✴pῗwo╴ ‘man, husband’), a final particle (fu2/✴bῗwo╴ ‘n'est-ce pas?’), and a word (also read (fu2/✴bῗwo╴) which of all the common pronouns and particles is perhaps the one which has most persistently defied all attempts at a usable grammatical description.

The story of Balai Pustaka begins in 1908, a memorable year in the history of modern Indonesia, for that was the year in which Boedi Oetomo, the first nationalist organization in Indonesia, was founded. In the same year the Governor-General of the Dutch East Indies set up a Commissie voor de Volkslectuur, a Committee for Popular Literature. And even though there was no visible connexion between the founding of Boedi Oetomo and the setting up of this Committee, there certainly was an indirect link. Both the foundation of Boedi Oetomo and the need for an institution for popular literature had something to do with the expansion of modern education in the colony.

The British Museum manuscript Or. 843 is the second part of Kitāb al-fisal fī 'l-milal by Ibn Hazm and includes several additions to the printed text. The longest has two sections and begins on folio 142 r., 1. 7. The first is quite new, is on medicine, and is purely theological, a defence of the healing art because the results are foreordained by God like all else. The second, beginning on folio 143 r., 1. 7, is longer than the print, adds kīmiyā to the title, and goes more into detail. The wording of the two texts differs widely. The world is a sphere (kura) of one substance which remains unchanged though its qualities change. Chickens eat dung, an unclean thing, but this feeds plants and trees, and the strictest Muslim has no scruple in eating corn, fowls, and fruit. To argue from analogy is wrong except within narrow limits. The sorcerer's claim to be able to turn a man into a donkey is absurd. Only angels and jinn can change their shapes.