Prev | Home | Next

General

The word epigraphy is derived from two Greek words viz., epi meaning 'on' or 'upon' and graphie meaning 'to write'. Thus the word epigraphy would mean the study of inscriptions or rather old inscriptions collectively. The word inscription which comes from Latin word inscribere which means 'to write on or upon' and denotes the general term for writings cut upon stone or metal - which forms the subject matter of epigraphy. Thus epigraphy is that part of archaeology which deals with the inscriptions engraved on stone, metal and other permanent materials like wood, shell, etc. Though engraving is the chief characteristic of an epigraph, there are some exceptions where old writings in ink on rocks, boulders, caves, etc., are also accepted as epigraphs. An inscription may also be prepared by embossing, i.e., scooping out the letters on the surface of a slab. A person who is engaged in the decipherment and interpretation of the epigraphs is called an epigraphist. Manuscripts, though some of them might be old, are not included in the field of epigraphy. However, manuscripts on palm-leaves, birch-bark, paper, etc., when associated with other excavated materials are occasionally taken into consideration. It may be pointed out that the real or important distinction between a manuscript and an epigraph as source-material is that while a manuscript is a copy of a copy of the original, an inscription is the original itself. Rarely do we find later copies of the earlier inscriptions. Some examples of this kind are noticed in Tamil Nadu where, while renovating the temples, the earlier inscriptions are re-written or the walls of these temples. In Karnataka also, for example, we fine later copies of earlier copper-plate records engraved on the walls of the Sankha-basti at Lakshmesvar in Dharwar District 1. As wooden tablet, pot-sherd or a copper-plate with writing in ink and dug out from an ancient site by the spade of an archaeologist is also considered as an epigraph for the purpose of study. The coins, medals and seals also contain old writings and the inscriptions on them are called by the ten legends. Though their study is valuable from the point of view of epigraphy strictly speaking, scholars include their study under the disciplines called numismatics and sigillography, i.e., seals and sealings. While the seal- legend is in the negative, the sealing gives its positive impression. Hundreds of sealings and comparatively smaller number of seals have been found in India.
Palaeography
Epigraphy or the study of inscriptions require a two-fold qualification of its votary, viz., ability to decipher the script and secondly the ability to interpret the language and contents of the epigraph. This brings us to the field of palaeography and linguistics. In the study of Indian epigraphy, palaeography forms an essential and important part and is concerned with the decipherment of the inscriptions without reference to the contents of the records. Some scholars understand the terms epigraphy and palaeography in slightly different ways. I.J. Gelb, for example, states that epigraphy and palaeography are more or less synonyms and are frequently interchangeable but he also tries to make a distinction, in good usage according to him, by saying that while an epigraphist is interested chiefly in inscriptions incised with a sharp tool or hard material such as stone, metal, wood, clay, etc., the palaeographer studies mainly manuscripts on skin, papyrus, or paper written in drawn or painted characters. Thus, according to Gelb, an epigraphist deals with older writings while a palaeographer is concerned with comparatively recent writings2 . Diringer also makes such a distinction between epigraphy and palaeography3. But here in India, we consider palaeography as an integral and important part of epigraphical studies, and, as already stated above, it concerns with the decipherment part only without reference to the contents of the epigraph.
In India, old inscriptions are written in many scripts and several languages which render the task of an epigraphist extremely difficult and complicated. For deciphering these inscriptions the knowledge of different scripts and their development is required while for understanding their contents the knowledge and study of the concerned languages is necessary. Between palaeography and linguistics comes what is called orthography which is concerned mainly with the peculiarities of spelling.
Importance of Writing
According to the famous Orientalist Breasted, the invention of writing has had a greater influence in uplifting the human race than any other intellectual achievement in the career of man4 . Great men like Carlyle and Kant thought that the invention of writing formed the real beginning of civilization. As language distinguishes man from animal, writing distinguishes a civilized man from a barbarian. We will be able to realise the importance of writing if only we try to imagine for a moment how difficult it is to manage without writing - without books, newspapers, etc. - in a cultured and civilized society. Our means of communications will be paralysed and what would happen to our knowledge if we cannot read about the achievements of the past. Since culture is defined by many scholars as 'communicable intelligence' and since writing is the most important means of communication, it becomes the main vehicle of man's culture and civilization.
A language appears to us, at any given moment as a stable structure of lexical and grammatical habits5 . But linguistic scholars have shown that this is an illusion and that every language is undergoing, at all times, a slow but unceasing process of linguistic change6 . This change can be ascertained only in the case of speech-communities which possess written records. The Kannad?a language of the 6th and 7th centuries A.D., of the time of Nripatunga (9th century A.D.), Pampa (10th century A.D.) and Ranna (11th century A.D.) is unlike the Kannad?a language of today. Similarly the English language of Chaucer and Shakespeare is unlike the English of today. Since written records give us direct information about the speech-habits of the past, the language phenomenon called linguistic change could be studied by a study of these written records only. Hence the written records are important for tracing the historical development of a language. Here the attention of the readers may be drawn to the important distinction between a manuscript and an epigraph as source materials of written records referred to above.
The relation between speech and writing is indeed very close. It is impossible to understand a writing without the knowledge of the speech represented by that writing. It is also difficult to study a speech without the knowledge of its writing." Writing is the graphic counterpart of the speech and it fixes the spoken language in a permanent form. It is the durable and visible representation of the language.
Learning the language, viz., one's mother tongue, is largely an unconscious process which is imitative and which goes on until control has become automatic. But reading and writing is started by teaching and comes later in life. Confusion between speech and writing exists even amongst educated persons. The distinction between sound and symbol - speech and writing or the letter and the sound it represents should be. clearly understood. The Kannad?a grammarian Ke#sira#ja who lived in the -13th century A.D. seems to make a reference, in his work Sabdaman$idarpan$a, to this distinction when he says, that the aksharas function in two ways, viz., sra#van$a 'what is heard by the ears' and cha#kshusa 'what is seen by the eyes'. According to structural linguists, the term 'language' should be used, in linguistic parlance, for spoken language or vocal language, i.e., communication by means of speech and the term 'written language' should be used for communication based on writing7. The difference between written language and spoken language is found in all the literary languages and writing has exercised much influence on the development of the languages. Written language preserves older forms of speech and has a restraining influence on the growth and development of speech. This is noticed while studying the traditional grammars of a language. If writing is wide-spread in a society, then the language does not change much. There is greater change in the language of the societies which are deprived of writing. The writing system also helps to create a sort of standard and unity in the written languages of a society whereas spoken languages exhibit a large number of dialectal varieties.
Writing obviously pre-supposes the existence of spoken language and is a relatively recent invention. The appearance of man in this world is believed to be about one. million years ago and the antiquity of language is also as old as man himself. But writing had its origin at some time in the comparatively recent past in the course of the intellectual development of mankind. Writing was thought of when man came to feel the need or necessity to communicate his ideas and thought to persons who are not present to hear him and to persons of distant places. Thus writing originated with the idea of finding out the means of transferring language. Though some cave-paintings and rock-paintings have been found as belonging to the Palaeolithic period (about 20,000 B.C.), scholars are of the opinion that the regular system of writing as commonly understood was introduced not earlier than the 4th millennium B.C. The oldest writings in the world are considered to be Sumerian, Egyptian, Proto-Elamite, Proto-Indic, Cretan, Hittle and Chinese. Their antiquity goes back to about 3000 B.C. and Gelb postulates a single common stream for the origin of these writing systems and calls it proto-Sumerian Pictographic system8 . He postulates the following stages of writing.
1. All writing originated in pictures. This is the most rudimentary stage of writing and is called pictogram. Picture-writing can be expressed orally in any language without altering the contents of the picture.
2. Ideographic Writing: In this, ideas are conveyed through picture, example, an expedition is shown by a number of pictures of men with arms in hand. This stage is called by Gelb as Semasiography or Forerunners of writing since real writing commences from this stage.
3. Word-writing or Logographic Writing: In this writing symbols are provided for words whose meaning does not lend itself by picture. Numerals (as denoted in an ancient Indian inscription) are examples of logographic writing. Also Roman symbols like X indicating the number ten are logographic. In this system, one has to learn a symbol for every word.
4. Syllabic Writing: In this, each symbol denotes one syllabic sound and has phonetic content. This is found in cuneiform and Egyptian writings.
5. Alphabetic Writing: In this each sound has a symbol indicating a vowel sound. All modern writings are alphabetic.
Thus in the long course of its history, writing progressed from stage to stage and showed uni-directional development. In this development according to Gelb, every writing system must pass through the differed stages of logography, syllabography and alphabetography in the succeeding order. And no writing can start with a later stage of writing unless it is a borrowed one, directly or indirectly, from another system of writing which has already gone through the previous stages. But the reverse development of writing from alphabetic system to syllabic system is not possible. However, a writing system can stop at a certain stage (like logographic system) without developing into syllabic system of writing9.
But this monogenesis theory of Gelb of all of the ancient scripts the world is not accepted by all scholars and, as will be shown below, will not hold good in the case of the history and development of Bra#hmi# script of India. Before discussing the origin of the Bra#hmi# script, let us turn our attention to a brief discussion about the discovery of and attempts made to decipher the Indus Script.

Indus Script
The world of scholars was totally ignorant about the culture known as Indus Valley Civilization or Harappa Culture till the early twenties of this century. The excavations at Mohenjodaro in Sind and at Harappa in Panjab (now in Pakistan) in 1922-23 and later and the discovery of numerous steatite seals in these excavations pushed back, at one stroke, the history of Indian Civilization including writing to the third millennium before Christ. After partition of India in 1947 when Mohenjodaro and Harappa went to Pakistan, similar sites in Eastern Panjab, Western Uttar Pradesh, Rajasthan and Gujarat were discovered. Of these Ka#liban#gan in Rajasthan and Lothal in Gujarat are important ones which have also yielded seals (and sealings) and have contributed much in establishing the chronological sequence of early and late phases of Harappan Culture. During the last fifty years and more, different views have been expressed by scholars about the authors of this great and highly developed culture which is comparable to that of Sumer, Babylon, Egypt and Assyria. And the key to the understanding of it lies in the decipherment of the writing on the seals and sealings. But unfortunately, the decipherment of this writing has defied the attempts of several scholars during the past fifty years and more. While some scholars think that this writing is of indigenous origin, others feel that it is of foreign origin. Even amongst those who think of indigenous origin, one set of scholars propound the theory of Dravidian origin while the other set put forth the theory of Aryan origin. These different views may be briefly mentioned here.
As stated above, the Indus script appears on a large number of steatite seals which are beautifully prepared. From Lothal in Gujarat some sealings are also found. From these sealings, which are found in association with packing material:; such as cloth, matting and twisted cords, it has been suggested that the sealings were used as labels and affixed to the packages of goods were thus of commercial or merchandise value10.
Waddel was one of the earliest scholars to attempt the decipherment of the Indus script. He thought that the writing represents the Sumerian script and, based on the identity of Sumerians with Aryans, he read the names of Vedic and Epic persons11 . Pran Nath assigned alphabetic values to the script and suggested their connection with the later Bra#hmi# script12. Sankaranand and Barua also thought that the script was alphabetic. Sudhansu Kumar Ray also held a similar view. Hrozny tried to connect the script with the Hittite language. But Heras suggested that the script is picto-phonographic and connected it with Dravidian languages reading old Tamil on the seals. Hevesy sought to establish similarity between the Indus script and the script of the Eastern Islands in the Pacific ocean. Hunter felt that the Indus script is derived -partly from Egyptian script and partly from Mesopotamian script. Flinders Petrie also connected the Indus script with the Egyptian hieroglyphs and thought that the seals contained only the titles and not the names of the officials. He also assumed that the symbols were ideographs while Meriggi thought that while symbols were ideograms, others were phonemes so that the writing was of ideo-phonographic system. Amongst other early scholars who have attempted to read the Indus script may be mentioned Gadd, Sydney Smith and Langdon13 . David Diringer remarks "it seems obvious that the Indus Valley script which is rather schematic and linear on the extant inscriptions, was originally pictographic but it is impossible to decide whether it was truly indigenous or imported"14.
The above discussion would show how scholars are holding different views regarding the Indus script and how difficult the problem of decipherment of this script has been during the last several years. It is possible to decipher an unknown language in a known script or a known language in an unknown script. But in regard to Indus script, it is a case of deciphering an unknown language in an unknown script and hence it has baffled and defied the attempts of many a well-known scriptologist. For the success of such an attempt certain points of contact are necessary15. For example, the script of the Egyptian hieroglyphs remained undeciphered for a very long period until the discovery of the famous 'Rosetta Stone' inscription in 1799 by the French engineer Bouchard at the time of Napoleon's expedition to Egypt. This sensational discovery proved to be a turning point in understanding the nature of the hieroglyphic script, because the Rosetta Stone contained inscriptions in three different kinds of script, viz., hieroglyphic, demotic (or local script) and Greek. With the help of the Greek text attempts to decipher the other two scripts were made by pioneers like the French orientalist Silvestre de Sacy and the Swedish diplomat Akerblad. And it was left to the fortune and credit of Sacy's pupil and French scholar J. Fr. Champollion to finally and conclusively decipher the inscriptions of the Rosetta Stone in 1821-2216. Again, the decipherment of the Cretan Linear B inscription, whose language was unknown and for which no bilingual text was available, was made possible for Ventris and Chadwick by the existence of, similar scripts n Cypriot and in Greek mainland written in Greek language17. Similarly some points of contact are necessary to find a satisfactory solution to the problem of the decipherment of the Indus script like the biscriptal or bilingual inscriptions.
We shall now review the recent attempts made by Indian and foreign scholars about the decipherment of the Indus or Harappan script. Amongst the foreign scholars, the Russian team consisting of Knorozov, Volcok and Gurov may be mentioned. They are credited to have taken the help of the computer machines. They assign word value to the signs and suggest that the script belonged to the Dravidian family of languages18. The Finnish team of scholars led by Asko Parpola also believed the language of the Indus script to be Dravidian. Amongst the Indian scholars I. Mahadevan and S.R. Rao have made a detailed study of the problem19. While the former is inclined to attribute the script to be Dravidian, the latter thinks it to be pre-Vedic. Mahadevan has also made use of the computer facilities and has attempted to achieve 'word-division' in the script assuming the language to be Dravidian. S.R. Rao claims that his approach is without any presumption and has tried to show that there has been a change in the script from its earlier phase to the later phase in that the number of signs which were more in the earlier period were reduced considerably in the later phase. He compares the signs of this later phase with the symbols of the North Semetic script of a comparative date and by showing similarity between them gives the same phonetic value to the Harappan script that is found in the Semetic script, other words, S.R. Rao suggests that there was evolution of the Indus script from an earlier period or mature peri9d (2500 B.C. to 1500 B.C.) and that the early syllabic-cum-alphabetic writing was disciplined into an alphabetic system by 1500 B.C. He also suggests that the Indus people spoke an Indo-European language which shows close affinity to Indo-Aryan in vocabulary, semantics and phonology. The names of the rulers and chiefs and of countries, sacrifices and divinities, as read by him, would suggest that the Harappans were the progenitors of the Vedic Aryans.
B.B. Lal has pointed out the difficulties in accepting the views of both I. Mahadevan and S.R. Rao20. Mahadevan himself has changed his views and methods of approach on more than one occasion and we have yet to wait and see what his final views in the matter of decipherment of the Harappan script are. In one of his latest papers entitled 'Study of the Indus Script : A Bilingual Approach'21 he has suggested that the problem should be studied from the point of view of interpreting the ideograms in the light of the Indian historical tradition which has come down to us in two main streams, viz., Indo-Aryan and Dravidian. This theory still remains to be tested by scholars before expressing any opinion.
As regards S.R. Rao's approach, viz., assigning the phonetic values of the Semitic script to the late Harappan script and reading it as pre-Vedic Sanskrit is also not decisive and final. As pointed out by Lal, Rao compares the symbols of the late Harappan with those of the Semitic ones and this material is not enough to arrive at any conclusion. Regarding the vowel sign Rao compares it not with any Semitic sign, since Semitic has no vowel signs, but with Sumerian sign for a following Waddel. And for some signs, Rao suggests different sources, viz., Akkadian and Ugaritic. This will be a difficult proposition. Moreover, while the late Harappan script, as suggested by Rao, has many vowel marks, the- Semitic script is completely devoid of any vowel marks. In justification of his approach, Rao says that he is proceeding from the known script (Semitic) to the unknown script (Harappan). But he is silent about the known origins of both these scripts which are different. While the Harappan is descended from the early Indus script according to Rao himself, the origin of the Semitic script is suggested to be Egyptian. So Rao has to explain at what stage the Semitic script acquired the phonetic values of the Indus script if his theory is to be supported. He has not attempted to answer these questions but has only instituted some comparison between the two scripts and has tried to establish some kind of phonetic relationship. Another defect in Rao's findings is that while he has given his reading as pre-Vedic or Indo-European, he has quoted not a single authority of Indo-European Linguistics or even an authority of Vedic language that the readings given by him can be accepted. In view of what is said above, it is not possible to accept Rao's claim that he has deciphered the Indus script. Of course, every scholar who makes an attempt at decipherment of a new script does claim that he has deciphered, but a new script can be taken as deciphered only when the world of scholars accept his views without any doubt22. So we can say, without any fear of contradiction, that the Indus script has defied the attempt of all scholars so far and has not yet been deciphered just as the Asokan Bra#hmhi# script has been deciphered. The discovery of longer record in the script or a bilingual or biscriptal writing or some definite contact point would help us in finding a satisfactory solution to this unknown script in an unknown language.

Antiquity of Writing in India
After the Indus script which covers the history of writing in India from about 2500 B.C. to 1500 B.C. and before the period of the Mauryan king Asoka (272 B.C. to 232 B.C.) whose edicts written on stone are available and which have been deciphered, the position regarding the history of writing in India has to be understood from references made in literature only. According to Diringer, the extensive Vedic literature does not refer to the existence of writing and the Vedic mantras were transmitted orally from mouth to mouth by the priestly class. This view is supported by Rhys Davids. According to these scholars, amongst ancient Indians while there is a goddess called Saraswati who represents goddess of learning, knowledge and eloquence, there is no god or goddess of writing. But this view is not accepted by other scholars. They say that the art of writing was known to the Vedic people. The Rigveda mentions the names of several metres like Ga#yatri#, Anusht?ubh, Br$ihati, Jagati, etc23. The Vaisaneyi Sam#hita#24 and the Atharva Ve#da25 also refer to the metres. The names of these metres and the technical terms regarding their composition involving rules of prosody presuppose the existence of writing. Again, the references in Vedic literature to the numerical figures like dasa (ten), sata (hundred), sahasra (thousand), niyuta (hundred-thousand), etc., show that calculation of high figures would have been made possible only with the written figures which shows the existence of the knowledge of writing. Such numerical figures and other calculations are found in the Bra#hman?a literature26 and also in Pa#n$ini's Asht?a#dhya#yi.
There are references to aksharas (indelible) and varn$as (coloured or painted) meaning written letters in the Upanishads . Similarly references to the words patra (leaf) and patta (a slab or tablet) indicate the practice of writing or engraving letters in ink and on hard objects. The discussion on sandhi or joining of letters and the composition of the sacred akshara o#m# as a combination of the letters a, u and m in Bra#hman$as also suggest the existence of writing. Thus many technical terms on grammar, etymology and prosody found in Vedic and later literature like the Upanishads, AÝranyakas and the Bra#hman$as suggest the existence of writing during that period.
The Asht?a#dhya#yi of Pa#n?ini (circa 4th century B.C.) contains the words lipi 'writing or script' and lipikara 'writer or scribe', yavana#ni Greek script' and grantha 'book', showing thereby that the knowledge of writing was known to Indians during that period. The Veda#n#gas and the Su#tra literature (circa 8th century to 3rd century B.C.) also certain references to writing. The Va#sishtha Dharma-Su#tra refers to the written documents28.
Kaut?ilya's Arthasa#stra (circa 4th century B.C., i.e., pre-Asokan period) states as follows:
(1) The student should learn after his tonsure ceremony, the alphabet and writing.
(2) The king should correspond with his ministers (patra sam#pre#shan?a).
(3) The king should send his spies with signs and writings.
(4) The writer should be an able person in reading documents, prompt in composing and elegant in writing.
From this, it is clear that writing was known during Kaut?ilya's period. The two epics, Ra#ma#yan?a and Maha#bha#rata, though assigned; to the pre-Mauryan period (i.e., to the 5th century B.C.), belong, in their present form, to the Gupta period, i.e., about the 4th century A.D. They mention terms like likh, le#kha, le#khana and le#khaka which testifies to the existence of the art of writing during the epic age.

The early Jaina and Buddhist literature also contain references to writing. The Jaina works Samava#ya#n#ga-su#tra (about 300 B.C.) and Pan?n?avan?a#-su#tra (about 168 B.C.) mention eighteen types of scripts which where in use at that time. The expression namo# bham#bhi#ya liviye 'salutation to the Bra#hmi# script' appears in the Bhagavati#-su#tra. The Buddhist works Maha#vastu and Lalitavistara (3rd century A.D.) also refer to the writing. While the former work mentions the names of thirty scripts prevalent at the time, the latter one states that Buddha as a boy went to the lipi-sa#la# 'a school where writing was taught'. The Ja#takas refer to phalaka 'wooden writing-board' and varn?aka 'wooden pen' as instruments for writing. The Vinayapit?aka commend the art of writing (le#khana) to the monks. Expressions like le#kha le#khana, likhati, le#khaka, akkhara, pan?n?o# (leaves'), suvarn?n?apat?t?a, etc., occur in early Buddhist literature.
Some foreigners who visited India during Mauryan and later periods have made references to the writing in India. The earliest historical contact with India is that of Greeks in the wake of the invasion of Alexander the Great. His general Nearchus states that the people of Panjab knew the art of preparing paper out of cotton and tattered clothes for Writing purposes30. Curtius, another Greek writer, refers to the tender bark of certain trees, probably of bhu#rjapatra which was used for writing31. Megasthenes, the Greek Ambassador to the court of the Maurya king Chandragupta, writes that milestones were fixed on the roads for the convenience of the travellers32.
The above discussion would show that the writing existed in India prior to the period of the Mauryan king Asoka. But, unfortunately we have not yet come across any material evidence like inscriptions to enable us to know about the nature of this writing. Any future discovery in this regard will be an event of great significance. Till then, we have to take it that the earliest datable epigraphs discovered and deciphered so far in India are the famous edicts of the Mauryan king Asoka. In other words, the history of decipherable epigraphy in India commences from the time of Asoka (272 to 232 B.C.). We shall, therefore, proceed to discuss the script which is called Mauryan Bra#hmi# or Asokan Bra#hmi# script by the scholars. This Bra#hmi# is the mother, as will be shown below, of all the alphabets of the Sanskritic and Dravidian languages which are prevalent in different parts of India today. It is also the origin of many alphabets of South-East Asia including Tibetan, Ceylonese, Burmese and Javanese.
FOOTNOTES
1. South Ind. Ins. Vol.XX, p.vii and Nos.3-7.
2. Gelb, A Study of Writing (1969), p.22.
3. Diringer, Writing, p.20.
4. J.H. Breasted, The Conquest of Civilization (1926), pp.53 ff.
5. L. Bloomfield, Language (1954), p.281.
6. Ibid.
7. Cf. H.A. Gleason, An Introduction to Descriptive Linguistics (1955), p,301.
8. Gelb, Study of Writing (1969), p.191.
9. Gelb, Ibid. (1969), pp.200-01.
10. Journ. Andhra Hist. Res. Soc. Vol.XXXIII, pp.1-2.
11. The Indo-Sumerian Seals Deciphered, pp.23-26.
12. "The Script of the Indus Valley Seals", JRAS, 1931, p.674.
13. The names of scholars and their views given here can be found in
(1) Dani, Indian Palaeography, pp.13-14 and
(2) T.V. Mahalingam, South Indian Palaeography, pp.91-96.
14. The Alphabet, p.85.
15. Cf. Dani, Indian Palaeography, p.14.
16. Hans Jensen, Sign, Symbol and Script (1970), pp.74-75.
17. Ibid., p.127.
18. Arlene R.K. Zide and Kamil Zvelebil (eds.) : The Soviet Decipherment of the Indus Valley Script, (Mouton, 1976).
19. Journal of Tamil Studies, Vol.II, No.1; Indian and Foreign Review, Vol.17, No.1.
20. Paper published by the Indian Institute of Advanced Study, Simla.
21. Presented at the Second International Conference on South Asian Languages and Linguists, Hyderabad (1980). Dr. E. Annamalai, Deputy Director, Central Institute Indian Languages, Mysore, kindly gave me a copy of this paper and my thanks are due to him.
22. For a further controversy on Mr. Rao's views, see, Deccan Herald (Daily) dated 16th September 1980, 20th September 1980 and 28th September 1980.
23. X, 14.16.
24. XI, 8.
25. VIII, 9.
26. Satapatha Bra#hman?a, XII, 3, 2, 1.
27. Chha#ndo#gya, I, 13 : II, 22, 3.
28. XVI, 10.14-15.
29. Ind. Ant., Vol.XXXIII, Appendix, p.5.
30. Ind. Ant., Vol.XXXIII, Appendix, p.6.
31. Mc Crindle, History of Alexander's Invasion of India, VIII. 9.
32. Indica, 91, 125-26.