According to the Prakrit Grammarians of the Western school, the Śaurasēnī-Mahārāṣtrī equivalent of Skr. ṣṭh is ṭṭh (Pischel, § 303), but the Māgadhī equivalent is sṭ (or, according to Namisādhu, śṭ) (Hc. iv, 290 ; Pischel, § 303). Nevertheless, the Māgadhī equivalent of tiṣṭhati is ciṣṭhadi (Hc. iv, 298), as compared with the Śaurasēnī-Mahārāṣṭṛi ciṭṭha (d) i. I shall deal later on with the initial c. So much for the Western school.

In the first part of this paper it was suggested that a cave was dedicated at Nasik either in or before the eighteenth year of the reign of Gautamïputra Śātakarṇi, and was given to people for whose benefit land previously enjoyed by Uṣavadāta was granted. Subsequently Śātakarṇi's mother, Gautamï Balaśrī, caused this cave to be enlarged by adding chambers which she claims to be her own benefaction. At that time I believed the veranda in cave No. 3 at Nasik to be the original cave dedicated by Gautamïputra Śātakarṇi and the remaining parts of the same cave to be the gift of his mother. During the last four years I had to visit Nasik repeatedly on duty and I have come to the conclusion that a portion of cave No. 3 of the Pandu Lena group, as it stands at present, must have been dedicated by Gautamïputra Śātakarṇi, and inscriptions No. 4 (Kshatrapa 13) and No. 5 (Kshatrapa 14) had been incised in the walls of this original cave before its enlargement by Queen Gautamī Balaśrī, when inscriptions Nos. 2 (Kshatrapa 18) and 3 (Kshatrapa 19) were incised on another wall.

The country where Tirāhī is spoken belongs to that border region between India and Eastern Irān which since my youth has never ceased to attract me by its varied historical, geographical, and linguistic interest. So when after my first Central-Asian expedition my hope of being employed in the newly constituted North-West Frontier Province was realized, I was anxious to use whatever chance might offer for securing materials concerning that old language of Tirāh, to the supposed survival of which Sir George Grierson had first called my attention. The inquiries I made with this object during 1904–5 both from the side of Peshāwar and from that of Kohāt and the Kurram valley proved, however, fruitless. The Political Agents of the Khyber and Kurram, whom their duties as Wardens of the Marches keep in close contact with the tribes occupying the valleys of Tirāh and the tracts of Afghānistān adjoining towards the Kābul river, uniformly asserted that their local informants knew of no other language but Paṣḥtō being spoken between the Kurram and Kābul rivers. Other Frontier Officers to whom I applied were equally unable to help. It seemed, indeed, as if that Tirāhī speech, of which Leech in 1838 had heard as a “ relict ” preserved by a small tribal community outside its original home, and had collected a very brief vocabulary, had since become completely submerged by the steadily spreading tide of Paṣḥtō. Nor did I fare better in 1912, when archæological work brought me back for some months to the Peshāwar border before my third Central-Asian expedition.

This study is based on the Weber-Macartney MS., which was edited by Hoernle, with a facsimile reproduction of the MS., transcription of the texts in Roman characters, and an index. (JASB., vol. lxx, pt. i, Extra Number 1, 1901.) Hoernle's text was pioneer work in the decipherment of Kuchean recently known as Tocharian B. Since then a few letters have been proved to have different values.

Whilst engaged as a Research Student on “Mediaeval Arabian Musical Theory ” at Glasgow University, I had the good fortune to have access to several libraries, and among them the Bodleian, where I compiled for my needs a short but detailed catalogue of the Arabic musical MSS. in this famous library. As this catalogue may be of use to orientalists and musical antiquaries, I take this opportunity of giving it publicity.

Tirāhī is a Dardic language of which, till now, hardly anything has been known. Our only materials have been the contents of a short vocabulary compiled by Leech in the year 1838. According to him, its speakers once inhabited the Tirāh country, now the home of the Afrīdi Pathāns, and, in consequence of a feud breaking out between the Ōrakzāīs and the Afrīdīs, they left that tract and settled in the Ningrahār country in independent Afghānistān, where they are now found.

More than thirty years ago Professor Sylvain Lévi, in his excellent book Le Théatre Indien (1890), p. 361, suggested that the Śakāra, the ridiculous, brawling scoundrel of a king's brother-in-law (śyāla) in the Hindu drama, is in reality a Śaka, a type of the Scythian princes ruling during nearly four centuries in Western India, as seen from the Indian point of view. This very judicious suggestion was repeated by Professor Lévi some ten years later in a very interesting article in the Journal Asiatique, ix, 19 (1902), p. 123, and might be looked upon as being fairly well established though its author has not given any detailed proofs to support his hypothesis. No reasons whatsoever have been adduced by Professor Konow in his book Das indische Drama, p. 15, for rejecting this theory, and so we need scarcely regard his rejection more seriously than several other less well-founded assertions in his work.

The existence of poets in Arabia before the rise of Islam is certified by the Qur'an, which contains one Surah named after them, and occasionally alludes to them elsewhere. Among the descriptions of the Prophet given by his opponents there was “ a Jinn-ridden poet” (xxxvii, 35), to which he replies that he has brought the truth. In another passage (lii, 29) the suggestions that he was a kāhin, a jinn-ridden man, and a poet are offered as alternatives. Since those who described him as a poet said they would wait to see what would happen to him (lii, 30) it might be inferred that poets were in the habit of foretelling the future. Elsewhere he asserts that his language is not that of a poet, but rather of an honourable messenger (lxix, 41), and that God had not taught him poetry, which would have been of no use to him (xxxvi, 69); his utterances were “ statement and clear lesson ”, whence we should infer that poetry was obscure. These hints about the poets are summarized in the Surah that bears their name (xxvi, 224, foll.), where we are told that they are followed by the misguided, rave in every valley, and say what they do not do. The sequel might seem to except certain pious bards from this condemnation, but the style of the Qur'an renders it uncertain whether this exception really applies to bards.

The following pages bring together passages from Chinese books, some of which have been translated before in a different context, illustrating them by reproductions of woodcuts which are unfamiliar and of great interest.

Meaning has been defined by the majority of ancient Indian writers on the philosophy of language in terms of a relation. Thus, the great grammarian Nāgeśa Bhaṭṭa defines meaning as a particular relation between the word and the object denoted. This relation is a power, which exists in the object as signifiability, and in the word as significativeness. It is only by the cognition of this relation that the presentation of objects by means of words is possible.

In the Centenary Supplement to this Journal, pp. 41–8, Mr. G. R. Driver edited a Neo-Babylonian tablet concerned with the sale of a GÍR.LAL-priesthood, and made some excellent observations on this class of transaction. In what follows there will be frequent occasion to refer to that article.

The selection of a title is sometimes a choice of evils. Which shall most move us, the desire to attract the many, or the fear of offending the few?

For the heading of this paper, the original aim had been laudable. Pith, brevity, and point were to combine with brilliant exactness of definition, alluring to the reader and doing justice to the theme. “ That,” as they say in the street, “ was the idea.” Followed an interval for intellectual refreshment and repose, and then, “ sudden in a minute,” in a flash, was captured the caption. But the disillusion of it! Gone were the pith and the brevity, and in their stead appeared a sesquipedalian pomposity. But however ponderously, the present title has the merit that it does precisely formulate the baffling and hitherto little noticed phenomena of substitutional types to which this paper essays to call attention by the illustration of a few out of many examples.

Nalōdaya, the short alliterative literary curio of four chapters, which is “ confessedly a difficult work as much so as Persius is in Latin and Pindar in Greek ”, and which is a superb example of the ingenuity displayed “ in the manipulation of varied forms of alliteration and elaborate tricks of style ”, was for a long time considered to be the work of that prince of Sanskrit poets, Kālidāsa, not because that there were any colophonian clues in the poem itself that could countenance such an attribution, but that the uncritical pandits of the orthodox set had opined that Kālidāsa alone could have been capable of the difficult feats of verbal gymnastics exhibited in every line of the kāvya, and had, as a salve to the literary conscience that perceived the incompatibility of such an artificial poem as the Nalōdaya emanating from the same hand as wrote the Raghuvaṃśa and the Śākuntala, invented the story that the poet was goaded on to its composition by the mischievous scepticism expressed by his friends as to his alliterative capacity. But since the discovery by Professor Peterson of a manuscript of the Nalōdaya whose commentator, Rāmarṣi (c. A.D. 1600), had attributed the work to one Ravidēva, scholars have been entertaining the view that this kāvya was the production of a certain Ravidēva, about whom no other biographical details were, however, available. I shall here examine the question of the real authorship of the poem in the light of the information furnished by another commentary on the same work, which is available in manuscript in the Palace Library of H.H. the Maharaja of Travancore.

Like the invasion of the Danes in England, the incursions of the Bargi in Bengal, were characterized by murder and rapine and every species of atrocity. There is, however, a good deal of difference between the two, for we have nothing comparable to the Anglo-Saxon chronicle to record the latter, and in the matter of other materials, we are sadly lacking. The Nagpur Marhattas have left no historical records. Consequently there are no Marhatta sources. Neither are there any letters in Marathi, at least to my knowledge, on the subject, as these raids were undertaken by the now defunct house of Nagpur. Mr. Hill in his Bengal in 1756–7, has given us a brief summary of the English Factory Records of Fort William, and the old records of the East India Company might, if thoroughly searched, yield good material. The time, however, demanded for this is beyond our power to afford. Besides the above and occasional references here and there, there are, of course, three ether books which can be profitably consulted for the purpose—Salimulla's Tarikh-i-Bangla, Riyazu-s-salatin, and the Seir Mutaherin. This practically exhausts our list of authorities for studying the subject, so far as they are available to the general student.

My lists of examples are taken mainly from Hübschmann's Armenische Grammatik, the lists in which were practically exhaustive in the state of knowledge at the time at which they were composed. Additional examples I have taken from Meillet, whose article in the Mémoires de la Société de Linguistiqtie for 1894 deals with this very problem. His Esquisse de la Grammaire de l'Arménien Classique, produced in 1903, also provides interesting information. What examples are not derived from these sources I have taken from Brugmann's Grundriss, and from certain other scattered articles, references to which appear as they are cited. To the three scholars mentioned the debt of any student of this difficult language is overwhelming.

Ḫarrian is one of “ the eight languages of Boghazköi ”, that spoken in the eastern part of the Hittite Empire. The principal facts concerning Ḫarrian may be found in two articles by Dr. Forrer in Mitteilungen d. deutschen Orient-Gesellschaft, 61, Dec, 1921, and in Zeitschr. d. deutschen Morgenländ.-Gesellschaft, 76, 1922, 171 ff. The articles will be referred to as MDOG. and ZDMG.

Two sannyāsī problems have long troubled me, first, how it was that the early rule, Any educated twice-born man can become a sannyāsī, was supplanted by the rule, Only Brāhmans can enter the order, and secondly, how it came about that, in North India, seven of the ten sub-orders of sannyāsīs became flooded with non-Brāhmans and lost their purity.

Attention has recently been called by Dr. Blackman to an important ritual text of Babylonian origin, of which the only versions hitherto known, from Ashurbanipal's library, were published by Professor Zimmern. The editor of these texts has pointed out their possible affinities, not only with the Egyptian ritual of “ Opening the Mouth ”, but also with Mandaitic ritual. Since the ritual appears to be of considerable importance for the study of comparative religion, attention must be called to another version in the British Museum, which serves to complete and explain what is already known to some extent. I am indebted to the Keeper of the Egyptian and Assyrian Antiquities for permission to publish this document.