Antiquities of Chamba State - An Old and Rare Book (Set of 2 Volumes)

When in the spring of 1902 my friend Mr. T. W. Arnold, then Professor of the Government College at Lahore, first drew. my attention to Chamba as possible field of antiquarian research, I little foresaw what a wealth of epigraphic records this petty Hill State would prove to contain. Sir A. Cunningham had visits Chamba in 1839 and devoted to it a few pages in two of his well-known volume In these he discussed the inscribed images of Meru-varman and the chief temples of the capital, gave a list of the Rajas and mentioned the existence at Chamba tow of four inscriptions, three on copper-plates and one on a stone slab. Here.tas elsewhere, the great pioneer of Indian archaeology only demarcated the field, leaving to others its further exploration.

During the years 1902-1908 I have been shle to devote part of the summer months to this pleasant task, and in the course of my tours have succeeded in recovering one hundred and thirty inscriptions. This result I attribute mainly to the vigorous support which from the outset my researches received from His Highness, Sir Bhuri Singh,K.C.S.I., C.I.E., Raja of Chamba, to whom, with the consent of the Government of India, this work is dedicated. The discovery of many a unknown inscription is due to the information supplied by: the officials acting under his orders; and I may mention here, that His Highness has further shown his interest in the antiquities and past records of his State by founding a local Museum, which was opened on the 14th September, ]908, and has been rightly named after him.

The importance of these inscriptions for local history. it is hardly necessary t emphasize. They do not, it is true, help us to solve any of the great problems c Indian history, but they enable us to write a more detailed and more coherent story of Chamba than of any of the other Himalayan States, excluding Kasmir and Nepal.

These records, moreover, throw side-lights on the history of neighbouring territories where ancient document, are less numerous and every scrap of information is of value. Even in places where historical documents do exist, their contents can often be supplemented from the Chamba inscriptions. I This is particularly the case in Kasmir-e-the only country in India of which we possess a written history of the Hindu period. The famous RajatarangiJ;li is replete with information about the 11th and 12th centuries, the same epoch to which the bulk of the Chamba inscriptions belongs. Thus it happens that Kalhana's chronicle imparts life to the stone slabs and metal plates of Chamba, whilst these contemporary documents, in their turn, confirm the trustworthiness of the great chronicler of Kasmir. Indeed, the excellent annotated edition of the Rajatarangini, both in Sanskrit and English, by Dr. M. A. Stein, C.I.E., has been one• of my chief guides in the course of my researches, and it has been no small satisfaction to me to continue in Chamba the work so splendidly inaugurated in Kasmir by that distinguished scholar.

There is one circumstance which lends to the antiquities of Chamba an interest far exceeding the narrow limits of local history. it is that the past to which they belong, is connected with the present by a tissue of unbroken threads. Whereas in other and more exposed parts of India one dynasty was quickly ousted by another, new creeds and customs came to supplant the old ones, and successive waves of foreign invasion swept away all remembrance of the past; Ohamba, engirdled by her snowclad mountain-barriers, has, century after century, retained ancient traditions and institutions, which are only now gradually giving way to the irresistible onslaught of Western civilization.

The antiquarian can hardly refrain from deploring this change which, though beneficial in many respects, so often tends to destroy that what is sanctified by the ages. But instead of lamenting the unavoidable changes of time, he will do well in retaining at least a record of the vanishing past.

Chamba is still ruled by a descendant of the noble house whose scions fought in the civil wars of Kasmireide by side with Harsa and Sussala. He still resides in the same "glorious Champa " whence his ancestors issued their copper charters, and where his subjects still hail him with the classical greeting" Jai Deo !" Up to the present day the people of Chamba worship in the temples founded by Sahillavarman in the l0th century, and at Brahmor, the ancient capital Brahmapura, the silent brazen idols are still enshrined in their wonderful wood-carved fanes in which Meru-varman piously placed them twelve centuries ago. The Ranas-those "barons of the Hills," whose former importance' was first revealed by the eulogies of Baijnath, are still met within Chamba, often, it is true, reduced to the state of poor peasants, but still clinging to their ancient title and to the ruins of their ancestral castles. The traveller too in the valleys of Chamba is still received after his day's march by an official whose title and presumably whose duties also have remained unchanged through the lapse of ages, though they have passed into oblivion in every other part of India.

There are among our inscriptions some which are distinguished by no small degree of literary ability such as the eulogies of Sarahan, Devt-ri-kothi and Mul-Kihar, But it must be admitted that the great majority do not attain a high standard' of scholarship, and we shall not be far wrong in assuming that, whatever erudition they display, was borrowed from the neighbouring seat of Sanskrit learning in Kasmir. The skilful poet of the Baijnath eulogy calls himself the son of a judge (p1'amala1') from that country, and it was perhaps the civil wars of Kalhana's days which compelled many a pandit to seek refuge with the rulers of the adjoining Hill States.

Though inferior in literary interest, the Chamba inscriptions are highly important from a palaeographical point of view; for they form an uninterrupted series of Sarada records ranging' from the time when this script was evolved out of the Western Gupta alphabet down to the Muhammadan period, when it developed into Gurmukhi, Takari, and 'other modern writings. In the course of the present work I shall endeavour to, show that Sarada was once extensively used both in the Plains and the Hills of the Panjab, and that, though this character was remarkably conservative, its forms were by no means so immutable as the best authorities on Indian palaeography have supposed. The Chambii records display a slow, but distinct development, and I feel confident that they will supply a reliable base for establishing the approximate date for any document written in this script.

It is true that the chronology of the Chamba epigraphs is attended with very great difficulty owing to the almost exclusive use of the Lokakala era. I have discussed the various thorny questions bearing on this subject in a special chapter of my introduction, but offer my conclusions with great diffidence, except where they could be. checked by so good an authority as the late Professor F. Kielhorn. His assistance-welcome always both for its own sake as for the kindness with which it was offered--I wish here gratefully to acknowledge.

Although in several cases I have not succeeded in fixing the precise date of the inscriptions here edited, it has nearly always been possible to arrive at approximate dates and to establish the order in which they succeed each' other. In this I have derived great benefit from the Chamba: VarhSavali or Genealogical Roll which His Highness has allowed me to edit in the present volume. This roll furnishes us with a fairly complete list of the rulers of Chamba from about A.D. 700 and its historical accuracy can be checked by the aid of the inscriptions. There is, therefore, much reason to assume that the ruling family of Chamba has indeed held sway in the Ravi valley for more than twelve centuries and may boast of an antiquity equalled by few reigning houses in India and none in Et: pe.

A few words must be added on the subject of transliteration. Each system has its defects and the use of diacritical marks gives transcribed texts per se an unpleasant look of artificiality and clumsiness. Whereas the advocates of one system reproach their opponents with the use of such" monstra " as Krsna, the latter may as rightly object to ungainly forms like Lichchhavi. I have, therefore, chosen to reproduce the texts in Nagari, which is preferred to transcripts in Roman .by most European and' all Indian Sanskritists: These texts, thus made accessible to indigenous scholars, will, it is hoped, stimulate their interest in the history of their own country.

In the introductory portions, where transliteration was unavoidable, I have followed the international system adopted by the Oriental Congress and by most Oriental Societies. At the time when the present work was written, this system was still followed in publications of the Archaeological Survey, though recently it has been replaced by the Anglo-Indian system. Only in the name dhamba I have retained the usual spelling; in all other Indian words the represents the non-aspirated hard palatal and the oh the aspirata.

In conclusion, I wish to give expression to my sincere gratitude for the manifold assistance received both in the discovery and in the study of the inscriptions of Chamba. The names of those to whom my thanks are due will be found mentioned in their proper places. Here I wish first of all to give expression to my gratitude for the cordial support which this publication has received throughout from Mr. J. H. Marshall, C.I.E., Director General of Archaeology in India. Particularly, I am indebted to Dr. J. Hutchison of the Church of Scotland Mission, who himself a resident of Chamba for the last thirty years, has ever been ready to supplement the evidence of the ancient records by his extensive knowledge of local customs, traditions and history. He has, moreover, rendered me a great service in reading through both the manuscript and the proofs of the present work. For the Tibetan inscriptions Included in this work I was able to depend on so good an authority as the Rev. A. H. Francke of the Moravian Mission.

I have much pleasure also in recording the great help which has been rendered to me by Pandit Thakur Das of Chamba whose services His Highness has kindly placed at my disposal whenever I visited his State. The Pandit's local knowledge, modesty and love for his native hills made him an ever-interesting and pleasant companion on my tours. He is one of the very few representatives of traditional Sanskrit learning in Chamba ; yet the study of the hist1'as has by no means closed his eye to the interests of "the World of the living." To 'the Pandits Daya Ram Sahni and Hirananda of the Archaeological• Department I owe some clever conjectures and useful references. Pandit Daya Ram has also assisted me in the tedious task of correcting the VamsAvali and in revising. the proofs.

Nor 'must I omit to mention the name of my photographer, Munshi Ghulam Nabi, who has accompanied me on many a hard march along the mountain roads of Chamba, He has taken all the photographs used to illustrate the present volume and prepared the estampages of the inscriptions here reproduced. The illustrations will testify to the quality of his work. The reproduction was entrusted to Messrs. W. Griggs & Sons, Peckham, London, and carried out with the care for which their establishment is rightly renowned.

The labour, both physical and mental, bestowed on collecting and deciphering these epigraphs, has been great. But" the labour we delight in physics pain.' And .. truly delightful has been the task of revealing the antiquarian treasures hidden in that glorious mountain region which a popular adage so rightly describes a! Chamiba achambha "Chamba the Charming."

Only three quarters of a century ago Chamba State was a terra incognita. The traveller, George Forster, who, on his journey through the Panjab Hills by way of Nurpur, Basohli and Jammu in 1783, almost skirted the western boundary of Chamba, refers to it as "the Chambay country a mountainous territory of large extent." This seems to be the first mention of the State by a European. William Moorcroft," in. describing the course of the Ravi, is more detailed. But he 'Only reproduces native accounts which had reached him on his march through Kangra in July 1820. Hence his information is confused and inaccurate. He notes that the Ravi in its upper course is called Raiva. It is curious that he takes the river of Manimahes and Harsar+-s.e., the BuQ.haJ-as the principal source. What is now considered the main river, he calls "the Siang from Bhaunso " (i.e. Bam Bhanso). He states correctly that the two branches meet at Ulans; but reverses the relative position of this place and Chamba. For the rest, the situation of Chamba at the junction of the Ravi and the Sava or Sal is correctly described.

He then makes the Tavi a tributary of the RaYi, and the Uj a tributary of the Tavi, whereas in reality the Tavi (on which Jammu' is situated) flows into the Cinab and the U j into the l{ii.yi. The confusion is probably due to the fact that the Tavi and the U j take their rise from the same mountain not far from Kund Kaplas. Koth Belota, mentioned by Moorcroft as the. place from which the Uj takes its origin, is probably the village Belota on the left bank of the Upptr Uj.

We have reason to regret that the distinguished traveller did not accept the invitation of the Rani of Chamba' who was 'indisposed and wanted his medical advice. He was apparently deterred-by the reports" about the difficulty of the- road thither, "it being necessary in many places to drag men and baggage by ropes up the scarps of the rock." Had Moorcroft personally visited Chantba, his account would no doubt have been far more accurate and detailed. The first European who visited Chamba was Yigne," His itinerary. though entertaining and full of interest, does not -possess the scholarly thoroughness of Moorcroft. He is less accurate in the rendering of proper names arul indulges in phantastical etymologies. Vigne visited Chamba in February 1839, in the reign of Raja Carhat Singh He came from. Basohl! and left again for Nurpur by the Cuari Pass. He gives a good account. of the capital, but did not visit the interior of the State.

It was only after the annexation of the Panjab that Chamba became Letter known to Europeans, especially as the abundance of game attracted many sport men. The antiquarian remains of Chamba were first brought to notice by Sir Alexander Cunningham, who visited Chamba for the first time in A.D. 18.39.' He extended his tour to the Upper Ravl Valley and was the first to describe' the ancient capital, Brahmor, and its temples.

It is now more than forty-five years since the monumental work entitled ANTIQUITIES OF CHAMBA STATE, Part I (Inscriptions of the Pre-Muhammadan Series. Its veteran author, Dr. J. PH. Vogel, the then Superintendent, Archaeological Survey, Northern Circle, had intended to publish the whole material in two parts, the first, as has been indicated above, dealing with the Pre-Muhammadan period, and the second covering the rest, namely, the Muhammadan and Post-Muhammadan periods. After having completed the first part, he had already started on the second, but had to leave it unaccomplished owing to his departure from India to Holland. Then set in the Great War which made any further progress with the work impossible. However, Dr. Vogel has all along been anxious to see that the work commenced by him is somehow or other finished, and to that end he had been negotiating with the Director General of Archaeology in India. And it was as a result thereof that I was entrusted with the task of writing the second part of the ANTIQUITIES OF CHAMBA STATE dealing with the inscriptions of and later periods. It has been a matter of great pride to me that I have been personally associated with the learned Professor, Dr. J. PH. Vogel, having worked under him for some years in the rooms of the Kern Institute at Leyden, Holland.

Dr. Vogel, as stated above, had commenced working on the present volume. When in the summer of 1938 Rao Bahadur K. N. Dikshit, the then Director General of Archaeology in India, and myself visited Chamba in order to take of the mate rial to be worked, we found inter alia Dr. Vogel’s own manuscripts, comprising well nigh three hundred pages, preserved in the Bhuri Singh Museum there. These were later on sent to the office of the Director General of Archaeology in India, New Delhi, and were subsequently placed at my disposal.

These manuscripts concern only the copper-plate charters which from the of this volume. They contain transcripts, in some cases complete and in the other partial, explanatory notes here, translations of parts of few inscriptions, identification of some of the places mentioned in the grants, and notes touching the plan of the work as envisaged by Dr. Vogel. All this material, it gives me pleasure to record, stood me in good stead; it afforded me both guidance and assistance in carrying out the allotted task more or less according to the plan visualized by Dr. Vogel himself.

The Bhuri Singh Museum at Chamba had in its possession also impressions of the great many of the copper-plate inscriptions dealt with here, which were likewise sent to the office of the Director General of Archaeology in India, New Delhi. A similar collection of impressions In the office of the Superintendent, Archaeological Survey, Frontier (then Northern) Circle, Lahore (now in Pakistan) which in like manner was made available. The two collections supplemented each other, though some few gaps still remained to be filled.

When the whole of the existing material was thus assembled in the office of the Director General of the existing material was thus assembled in the office of the Director General OF Archaeology in India, New Delhi, two post-graduates, namely Mr. S. K. Dikshit, M. A., and Mr. Krishna Deva, M.A., availed themselves of the opportunity of studying it. Besides checking it, they added some notes of their own.

It was early in the year 1939 that I was formally called upon to undertake the task and to visit Chmba in that connection. Mr. Krishna Deva was then deputed to accompany me for rendering assistance in the work.

A personal visit to Chamba was essential for various reasons. The eighty odd copper-plate charters of here are substantially couched in the Bhasha or the local dialect called Chambyali in a from which, in consonance with the nature of the subject-matter, necessarily differs from the common parlance and abounds in obscure expressions and abstruse terms, and, on top of that, is more or less obsolete at present. Such hard nuts could be cracked only with help locally available. Secondly, it was possible readily to identify places, rivers, mountains and so forth, that are mentioned in the records, likewise by personal enquiries. Thirdly, if any of the original documents were required either for examination or for comparison, they were easy of access only within the State. Besides, there was a possibility of discovering additional epigraphs.

Speaking of additional discoveries, it may be pointed out that three of the copperplate inscriptions dealt with here, namely Nos. 33, 81 and 82, have not been noticed by Dr. Vogel. Nor do the two collections of impressions referred to above include any impressions of them. They thus appear to be subsequent acquisitions. The first of them, that is No. 33, pertains to Balabhadra, while the remaining two, Nos. 81 and 82, belong to Srisimha. It is of course, quite likely that Dr. Vogel was aware of the existence of these last two, but that he considered them to be of too later a date be included in his collection. There is, however, one consideration that militates against such a supposition. The last copper-plate charter by him is said to be dated V.S. 1941, which is not included here. The two records in (Nos. 81 and 82) are earlier than that by one quarter of a century. On the other hand, it can be said, at least of the last record (No.82), that, build as it is in the right wall of the main entrance leading to the to the temple of Lakshmi-Narayana and other divinities, it is so prominently situated that it can hardly escape notice, even of a casual visitor there.

As indicated above, estampages of some of the copper-plate charters were still wanting. Since necessary facilities were to be had in Chamba for taking the required inked impressions of such records, we had to be content only with their photographs which were prepared by Mr. Mangat Rai Mehta, then attached to the office of the Superintendent, Archaeological Survey, Frontier Circle, Lahore.

During my stay in Chamba, which lasted for two months, much spadework was done: fresh transcripts prepared, elucidative notes taken and rough translation drawn up, all this with the help so generously lent by the Rajaguru, Pandit Thakur Das, who had formerly assisted Dr. Vogel in like manner, and to whose ability and worthiness Dr. Vogel has paid a well-deserved tribute of praise while acknowledging his assistance. In fact, he is admittedly the only person alive in the Chamba State wno can understand and interpret the language of the old documents edited here; and it is a matter of deep gratification that his services have still been available.

Before closing this note, I which to record my grateful acknowledgments for the help received from different quarters in the production of this work. First of all, my sincere thanks are due to Professor Dr. J. PH. Vogel and the late Rao Bahadur K. N. Dikshit, both of whow from time favoured me with various useful suggestions and encouraging remarks while the work was under preparation. Next, I wish to express my gratitude to the them Council of Administration, Chamba State, consisting of Col. H. S. Strong, C. I. E., the President, Dewan Bahadur Madho Ram, the Minister, and Sahib Har Govind, the Judicial Member, who readily and willingly provided me with all facilities in connection with my study of the inscriptions during my stay in Chamba and also evinced keen interest therein.

To the Rajaguru, Pandi Thakur Das, I am greatly indebted for his ungrudging help which has proved most valuable and indispensable for a proper treatment of the records concerned. Further, I am thankful to Mr. Jaiwant Ram, B.A., B.T., the then Head Master, State High School , Chamba, and Curator, Bhuri Singh Museum Chamba, who likewise lent me a helping hand in solving some of the knotty textual problems.

Last but not least, I have to acknowledge equally the assistance received from Mr. Krishna Deva, M.A., now a Superintendent in the Department. His collaboration at that stage has helped in expediting the work.

Completion of the task fills me with joy, amply compensating for the pains has cost me: Klesah phalena hi punarnavatam vidhatte!

The total number of copper-plate inscriptions that are included in this work is eighty-two. The first of them is of them is dated V. S. 1387, AND THE LAST v. s. 1915. They pertain to twelve successive rulers, from Vairisimhavarman to Srimha, as detailed below:-

It will be seen that literally more than half of the total number of three charters belong to Balabhadra alone. This striking disproportion testifies to the excessively generous dispositions of the prince, of which we shall by and by have more evidence.

Common Features of All Plates
Some characteristics are common to all the copper-plate charters discovered in the Chamba State and are peculiar to them inasmuch as they are not commonly met with in similar contemporary documents found elsewhere in India. They may be summed up as follows: (1) Every charter consists of a single sheet of copper, though the size varies greatly. (2) Each plate is provided with a handle to its proper right. The thus puts on the appearance of a takhti or a wooden board used by school children. In some cases this handle has broken away, partly or altogether. Again, some of the handles have a hole pierced in the centre through which may be passed a cord by means of which the plate could be hung on to a peg in a safe corner-that is what the owners of the plates probably used to do. (3) Every one of the plates is invariably en engraved only on one side. The predilection for such an arrangement is very much in evidence on certain plates, where the text runs on into excepting the handle, while the size of letters in its concluding portion is gradually diminished into the bargain. A typical example of this kind is supplied by the inscription No. 32, though No. 23 beats it in respect of congestion. (4) Almost every plate has a seal engraved usually in the top left corner in the shape of a rosette or some other ornamental design. The space in its centre is occupied by a legend invariably in Nagari characters, containing the name of the king to whom the deed concerned pertains. In a few instances the seal is replaced by the word sahi likewise characters. The significance of this has been discussed below.

Lithic Records
The stone inscriptions sealt with in this volume do not compare favourably-neither in quantity nor in quality-with the copper-plate records. Their number amounts to twenty-one. The most outstanding of the lithic record is the Chabutra stone inscription of V. S. 1717. It is the precise in it that vests it with importance, which has been duly considered elsewhere.

The nineteen short epigraphs of the Vajresvari temple are mere labls giving names of the artisans, except one which is dated and informs us that the (reparation) work was started on the given date. Miscellaneous Inscriptions
The short inscription appearing on the massive bell suspended from the ceiling of the mandapa of the Chamunda temple stands out by itself, as it cannot be included in either of the preceding two classes. It is dated, givens the names of the dedicator and the manufacturer of the bell, and mentions its weight and cost.

Palaeography
Out inscriptions represents two distinct characters: Devanagari and Devasesha. The use of the former is restricted mostly to the legends appearing on the seals of the copper-plate charters. Later on, however, Devanagari alone is employed in such records. The earliest of this kind in Chamba is Umedasimha’s grant of V. S. 1805 (No. 78).

Devasesha refers to a later development of the Sarada alphabet. Dr. Vogel has made an exhaustive study of the subject and has aeeived at definite conclusions which may, with advantage, be recapitulated hare. The evolutions of the scripts in question is, in brief this: Brahmi-Western Gupta-Kutila-Sarada-Devasesha –Takari. Kutila, of which Sarada is be the immediate descendant, continued up to the of the end of the 8th or the beginning of the 9th century A. D. which thus becomes the epoch of Sarada. The script continues undergoing slow but sure changes until, by the beginning of the 13th century A. D., its appearance is sufficiently altered to justify a separate desingnation. That is then Devasesha. Dr. Vogel has pointed out that scholars like Buhler applied the tern Sarada even to this later phase of the character, though distinguishing it as ‘later Sarada, but that a specisl name was desirable. Although the term Devasesha is but little know outside Chamba, yet it has been adopted for the sake of convenience.

Dr. Vogel was concerned mainly with the Sarada characters. He has fully described the formation of each individual latter. In doing so, he has drawn attention to the peculiarities of Devasesha as well, illustrating the points by referring to the two inscriptions written in that alphabet, facsimiles of which had ny then been published, namely the spurious Sai copper-plate inscription of Vidagdhavarman and copper-plate grant of Bahadur Singh of Kullu. These records belong to the 16th century A. D.

The inscriptions edited here are fairly numerous and range in data from the early part of the 14th century to the middle of the 19th century. As such, they afford us ample scope for studying the peculiarities of the character and its gradual development in the course of over five hundred years with greater precision.

Before proceeding with a detailed examination of this nature, I may point out that the original naiheads or wedges of Kutila, which turned into small horizontal strokes in Sarada, in turn, developed into top Strokes in Devasesha and lend it a distinctive appearance, Devanagari as we know, represents the climax of this process. In Devasesha some letters, like gh, th, n, p, m, sh, and s, still appear with an open top, while in Devanagari they are provided with a top stroke.

Initial Vowels
“In Devasesha initial a and a preserve essentially the same shape as in Srada, but the top is closed by means of a top stroke as in Devanagari, and the wedge at the foot of the vertical sometimeds becomes a triangular loop.” To be exact, the foregoing remark applies to a only (1). The length is denoted by a book instead of a triangular loop at the foot (2). Examples of the latter are comparatively rare. The triangular loop of a may be taken as a regular feature of the sign. The instances, as in No. 72, line 12, are only ephemeral. Examples of a may be seen in almost every inscription, while those of a are met with in No. 14, line 3; No. 22, line 8; No. 37, line 14, etc. The from of a in No. 1, line 5, is rather unusual.

Initial I has retained its original shape, which consists dots and a curve below (3). For examples, see No. 12, lines 7 and 10; and No. 37, line 10 where it occurs thrice. In certain cases, the two dots are replaced by two small circles (4), instances of which may be seen in No. 5, lines 15 and 16; No. 8, line 15. The sign of i in No. 1, line 16, appears irregular inasmuch as it is inclined toward the right. The sign in No. 76, lines 8 and 10, exhibit the mark of medial o instead of two dots or circles, though the same inscription in lines 7 and 8 shows the first sign as well. The sign with the superscribed medial stroke instead of two dots or circles appcars also in No.67, lines 8, 9, 10, 11, 12, 14, 15, etc. Thus, this constitutes a regular variety of the initial i.

The sign of initial I is of rare occurrence, but it presents a very interesting phase in its development. Discrussing its from in Sarada, Dr. Vogel observes that “the upper and lower dots of ancient sign, which consisted of four dots, have been converted into a vertical stroke with a wedge on each end.” This is exemplified by the Saraharr prasati, where it occurs only once (5). In Devasesha its development is remarkable and quite consistent with the peculiarities already noticed. In the case of initial a, we observed how the wedge at the foot of the vertical becomes a triangular loop. We have also observed that a wedge at the top of letters in general becomes a horizontal topstroke. This process has taken place in the case of initial I, with the result that the vertical stroke with a wedge on each end has assumed the from of the letter ra, (6). In the disposal of the two dots also, the writers of Devasesha have introduced a modification: they have removed them from the flanking position and placed them on the top. Thus the form of the letter appears as (7). Instances of this may be seen in No. 38, lines 8 and 10; No. 39, line 11.

The sign of initial u has not differed much in shape from its original (8). The observation made by Dr. Vogel concerning this good: “The upward stroke differentiates the u from t. In the later inscriptions the two aksharas are often hard to distinguish.” The upward tendency of the stroke (9) is, in some instances, carried to the extreme, so that it curls over the body of the latter. See, for instance, in No. 1, lines 11, 13, (10) and 17. Instances of the normal sign are abundant.