खिलेषु काशीकरः

Introductory

The R̥gveda Saṁhitā has, according to the Sakala śākhā, 1017 hymns. According to the Caraṇavyūha, the R̥gveda Saṁhitā has come down in five recensions: Sākala, Başkala, Āśvalāyana, Sārkhāyana and Mândūkāyana. (Besides these five, many more Sakhas (21, 25 or 27) are said to have existed.) Of these, the Sākala recension is well known and has come down in a definite form. (2. It is to be noted that Kausitaki is different from Sānkhāyana. A Kausitaki Ghya Sūtra as different from the Sānkhāyana Gțhya has recently been published (Dr. T. R, Chintamani, Madras, 1944 ) and the publication of other Kauşitaki texts is also promised.)

The other recensions did not differ much from the Sākala recension except an omission or addition of certain hymns or verses within or without the Mandalas and also within or without the hymns, as well as certain changes in the order of hymns or verses. No definite conclusions seem to have been drawn as to the exact nature of each of these recensions for want of decisive data.

The khilas

The additional hymns or verses called Khilani have come down traditionally even though it has not been possible to attribute them to definite recensions and it is customary to publish them as a supplement to the R̥gveda Samhita in the Sakala recension. Thus Prof. Max Müller has published at the end of his great edition of R̥gveda Saṁhitā with the commentary of Sayanācārya, 32 Khilas with or without accents which he found in certain MSS. of R̥gveda. (3. Second edition, Oxford 1890-2. ) Prof. Theodor Aufrecht in his edition of R̥gveda Samhita printed in Roman characters has published 25 Khilas some of which are different from those in Max Müller’s edition. (4. Second edition, Bonn 1877.) The Bombay edition of R̥gveda Saṁhitā (Saka 1810-12) with Sāyaṇabhāşya edited by the two Mahāmabopādhyāyas Rajaramasastri Bodas and Sivaramasastri Gore gives a good number of Khilas including the Nivids, Praişas, Purorucas, Kuntāpas etc. under the title Parisistāni. The Nirnayasagar edition of R̥gveda Samhita (Bombay 1891) also gave the Khilas. Prof. Peterson in his Catalogue of MSS. in the Alwar State Library has reproduced a few, though not all, Khilas with accents found in certain Saṁhitā and Pada MSS. of R̥gveda deposited in that Library. Recently Pandit S. D. Satavalekar of Aundh (Dist. Satara) has published in his edition of R̥gveda Samhita (Aundh, 1940) 36 Khilas, some of which are not found in the abovementioned editions ether fully or in part. Pandit Satavalekar, however, has not mentioned the MSS. material on which he has based his edition of Khilas, at least the additional portion of them. In all these editions the places where the particular Khilas are to be recited are also denoted, and in most cases they agree with each other. (1. He has kindly lent me two MSS. of Sankhāyana Samhita of R̥gveda which he must have used. See p. 894. He has, however, obviously used some additional material. Pandit Anant Yajñeswar Dhupkar Sastri of Marcell Goa (Portuguese India ) who worked for the second edition of Aundh R̥gveda, informs me by letter that for editing the Khilas he used a MS. of his own, about 200 years old. For two Khilas inserted in the Aundh odition by conjecture, see notes to Khilas II. 1 ; IV.9.)

By far the most important and extensive collection of Khilas is preserved in a MS. of R̥gveda Sarhitā discovered in Kashmir by Prof. G. Bühler and described for the first time in his “Detailed Report of a tour in search of Sanskrit MSS.”, Bombay 1877, pp. 35-6. The MIS. was procured for the library of the Deccan College, Poona, which is now transferred to the Bhandarkar Oriental Rescarch Institute, Poona, where it is deposited at present. The MS. was transcribed by Dr. H. Wenzel and the transcript was examined by Max Müller, Macdonell and other scholars in Europe. Macdonell made use of this transcript in his cdition of Brhad-devata (Harvard Oriental Series Vol. 5 & 6, 1904) and while describing in his introduction the Khila portions in the MS. he remarked that there seems reason to believe that a satisfactory edition of the text of these Khilas could be produced from the Kashmir MS." (p. xxiii). The credit of editing the Kashmir Khilas and bringing out their critical value goes to Dr. phil. J. Scheftelowitz who prepared and published in Roman characters a critical edition of the same at Bonn in 1906 under the title “Apokryphon des R̥gveda.” Since the Kashmir MS. gives a comprehensive collection of authentic Khilas, I propose to present here all Khilas in the Kashmir MS. with a comparative study of those published in other editions and also of those available in MSS.

The Kashmir MS.

The MS. is written on birch bark and consists of 191 leaves each measuring 10.25" by 9.5". The birch bark being very thin, cach leaf of the MS. is made up of two sheets pasted back to back so as to make it rather thick and serviceable for writing purposes. The leaves are written on both sides and are arranged in the form of a modern book. Thirty two lines on an average are written on each page and there are 40 letters in a line. Several leaves are much injured. Borders of the first and last few leaves have worn out. The MS. is written in Sārada characters and the handwriting is bold, legible and correct. (2. Cf. Catalogue of the Government Collections of Manuscripts, Deccan College, Poona 1916, MS. No. 1, Scheftelowitz, pp. 32-3,) The colophon of the MS. runs thus:- इति श्रीऋग्वेदे शाकलके शाखायां दशममण्डले ऋग्वेदः खिल-सहित-संहितारण्यसहितश् च संपूर्णस्, समाप्तं। सं ५१ भा शु ति १३ लिखितं । भट्ट-भीम-स्वामिनो रामि-स्वामिनः पुत्रः शबलस्वामिनः पौत्रस् सम्पाद्यतं समाप्तम्॥ शुभमस्तु॥. Thus the copyist of the MS. is the son of भट्टभीमस्वामिन रामिस्वामिन् and grandson of शबलस्वामिन् . According to A. B. Keith, these concluding words are written by a later hand and probably denote an owner of the MS. He also adds that the same hand has here and there made some corrections in the text. According to Bühler the MS. is 400 years old. The century in the date given in the colophon is lost, but Bühler adds that [Fara ] f 49 corresponds to A.D. 75. According to Keith 49 most probably corresponds to A.D. 1575. The only alternatives are A.D. 1475 and 1675, but he doubts whether the MS. is so old as the former or so modern as the latter of these dates.

In leaves 1-176 R̥gveda Saṁhitā is written along with the Sarvānukramaṇi of Kātyāyana. The Adhyāya system is followed in the enumeration of the Saṁhitā, the whole R̥gveda being divided into 64 Adhyāyas. There is also the Mandala division, but the Aştaka division is totally absent. At the end of each Mandala there is a sentence indicating the conclusion of the last and the commencement of the next Mandala. There is no division of the Anuvākas nor that of the Suktas ; the Adhyāyas are divided into Vargas, but the number of verse is not shown. This method often makes it impossible to know where a hymn begins or ends. The end of a verse or half verse is denoted either by a perpendicular stroke or by the absence of Sandhi where in fact it is expected according to the Sandhi rules. Like the R̥gveda Saṁhitā, the Sarvānukramaṇī is also divided into Adhyāyas and the Anukramaṇi portion belonging to each Adhyāya is given at the beginning of the same. The MS. belongs to the Sakala sākhā as is clearly stated in the concluding words of Mandala X and at the end of the whole MS.

  1. JRAS 1907, p. 225.

The Khila Portion

The Khila portion is contained in folios 176 Bm 189B of the MS.2 All Khilas are divided into five Adhyāyas, each Adhyāya again being divided into Vargas. Scheftelowitz has shown this Varga number while I have dropped it. Yet there was no clue to know where a Khila began or ended. The Khilānukramanī helped in this respect. At the beginning of each Adhyāya there is the Khilanukramani portion belonging to it. It records the number of verses in a Khila, its author, metre and deity. Thus we are in a position to reproduce the Khilas at least as they stood in the time of the Khilānukramanī. The information of the Khilānukramaṇi agrees with that in the Brahmanas, R̥kprātiśākhya, Brhaddevatā, R̥gvidhāna, Srautasūtras and Grhyasūtras. The MS. quotes at the end of each Khila the Pratīka of the next R̥gvedic verse before which the particular Khila is to be recited. If a number of Khilas stand together, the Pratīka stands at the end of the last Khila.

(2. At the end of the Khilas there is some portion of the Aitareyāranyaka (folios 189B. 191). It is divided into three Adhyāyas, the first two of which form the third Aranyaka. The third Adhyâya giving Sanhitāraṇyam is reproduced by Scbeftelowitz in his edition.)

The copyist of this MS. did not write the Samhita or the Khilas by mere memory ; he probably copied from a text written in Sārada script, as is clear from the several writing mistakes. (1. For the orthographical peculiarities of this MS., see Scheftelowitz, pp. 47-8.)

Accent

The text is accented throughout. The method of marking accents is peculiar. Generally in Vedic MSS. the Anudātta and Svarita are denoted by horizontal and vertical strokes respectively, while in this MS. only the Udātta and primary Syarita are marked. The Udātta is marked by a small vertical stroke. This kind of marking Udātta is also found in the Maitrāyani Saṁhitā (Cf. Schröder, Preface p. xXx). The primary Svarita is denoted by the sign ᳠. This sign occurs twice in the Khila portion (1. 11. 4 ; 12. 7).

In Kāțhaka Saṁhitā a strong curve above the line comes as a sign of the primary Svarita, while in Maitrāyaṇi Saṁhitā the curve lies below the line (Cf. Schröder, ibid.). If the next syllable is accented, the primary Svarita is denoted by ३ (Trikampa) even though the preceding syllable is short. This ‘३’ which stands after the Svarita syllable has either a horizontal stroke below the line or a perpendicular stroke above the line. Scheftclowitz has retained this method of accentuation in his edition ; the same has been followed here too, In the case of certain supplementary hymns and verses also he has mostly marked the Udātta. I have given all of them without accents. In certain editions and MSS. these later Khilas are found accented, but the accents are not unanimous.(5) It is to be noted that Max Müller has, in his edition, printed all such Khilas without accents.

Additional Material

For the constitution of the text of thc Khilas, I have also used two MSS. of Sankhāyana sākha of R̥gveda kindly lent to me by Pandit S. D. Satavalekar of Aundh. They come from Junagadh and Navanagar in Kathiawar. The Junagadh MS. was written in Samwat 1754-5 and the accents were marked in red ink in Samwat 1755-8. The colophon at the end of Aştaka VIII runs

संवत् १७५४ वर्षे भाद्रपद शुदि ५ बुधे श्रीजीर्ण-दुर्गनिवासीय आभ्यन्तर-नागर-ज्ञातीय पंडा श्री गोविंदसुत् पंडा श्री ॥ भवानीदास सुत रघुनाथेन लिखितं ॥ सं. १७५८ वर्षे कार्तिक शुदि २ बुधे पं । श्री ॥ भवानीदाससुत रघुनाथेन स्व० ॥

The Navanagar MS. was written in Samwat 1894–5 and the accents were marked in red in Samvat 1898. The colophon at the end of Aştaka VIII runs

संवत् १८९४ ना मार्गशिर्ष शुदि ११ प्रारंभः सं। १८९५ ना कार्तिक शुक्ल १२ भोमे अद्य श्रीनवा-नगर-वास्तव्य आभ्यंतर-वृद्ध-नागर-ज्ञातीय आचार्य श्री ५ भवांनसुत् गोविंदजियेन पंड्या श्री ५ जगजीवनसुत् रणछोडजीसुत् मोतिरांमने कृष्णार्पण लिखिदत्वा……

The text in both the MSS. is divided into 64 Adhyāyas. There is no Aştaka division, while the beginning and the end of a Mandala are mentioned. On comparison, I find that both these MSS. are exactly identical with each other except a few scribal errors. So I have preferred to refer only to the Junagadh MS. by the designation जुना, on account of its older age and its peculiarity of marking accents.

The accents in the Navanagar MS. are marked in the usual manner, i.e. the Anudātta by a horizontal stroke below the letter and the Syarita by a perpendicular stroke above the letter. In the Junagadh MS., however, the Svarita is marked by a horizontal line through the body of the letter and the Anudātta is marked as usual. (1. It may, by the way, be noted that both these MSS, ma ‘१’ instead of ‘३’ in three places (Val. I. 3; VI. 7; VII, 1) in the Vālakhilya hymns.) This method is exactly the same as that in part of one of the Alwar MSS. of R̥gveda Saṁhitā which are described in the Catalogue as belonging to the Šānkhāyana sākhā (See ‘A Second Report of operations in search of Sanskrit MSS. in the Bombay Circle’, 1883–84 by Peter Peterson; Extra No. JBBRAS 1884). There is no direct proof in the MSS. themselves showing their Sankhāyana character, just as there is none also in the Alwar MSS. One does not know what was the position of the Khilas found in these MSS. in the Sankhāyana sākhā. It is to be noted that most of the Khilas are included within Varga numbers in both the Alwar and Junagadh MSS.

The peculiarity of the Sankhāyana redaction as denoted in the commentary on Caraṇavyūha is found in these MSS., namely that the 10th Valakhilya hymn (VIII. 58) is absent and two of its three verses are inserted after ķV. X. 88. 18 as a Khila. Whether this peculiarity exists in the Alwar MSS., is not known from Peterson’s description. An exhaustive comparison of the Junagadh MS. with the Alwar MSS. could not be made, since the latter were not available on loan. Arrangements to obtain the required information could not also prove successful. I have, therefore, to rely on Peterson’s description of the MSS. and the text of some of the Khilas given by him.

A comparison of the Junagadh and Alwar MSS. reveals certain differences in the Khila portion :

  • (1) The Khila सूक्तान्ते तृणान्य् अग्नाव् अरण्ये comes in Alwar MS. after RV. V. 49; in the Junagadh MS. it comes after RV. VI. 48.
  • (2) The Alwar MS. does not contain Srīsūkta (Khila II. 6) which is present in the Junagadh MS.
  • (3) The Sivasamkalpasūkta (Yenedam IV. 11) which seems to be absent in the Alwar MS., is found in the Junagadh MS.
  • (4) Khilas III 19; 20; IV. 12 are found in the Junagadh MS.; they do not exist in the Alwar MS.
  • (5) In three Khilas the number of verses in both the MSS. is different : (
    • (a) Khila II. 12 has two verses in the Alwar MS., 3 in the Junagadh MS.
    • (b) In Khila II. 13 the Junagadh MS. gives 3 additional verses after the first 3 verses.
    • (c) In Khila IV. 2 (Rātrīsākta) the Junagadh MS. reveals two more verses than the Alwar MS., one of which is found in the Nirnayasagar edition of R̥gveda and the other is quite a new one.

Barring these differences, the Junagadh and Alwar MSS. agree with each other. One is inclined to gather from these differences in Saikhāyana MSS., that the Khilas were probably not enumerated as a part of the Samhită in spite of their enumeration within the Varga number. The question, however, remains, which Khilas were recognised as Khilas in the Sāňkhāyana sākhā. Peterson has described the Alwar MSS. in comparison to Aufrecht’s edition of Khilas, and one is not sure about the position of the Alwar MSS. with regard to the Khilas not found in Aufrecht. The end of the Khila portion is exactly the same in both MSS. After giving the Khila विदा मघवन् विदा, both the MSS. have अग्निदेवेद्धः विदा मघवन्विदों. The text of the Khilas in the Junagadh MS. is mostly identical with that in the Alwar MSS. I have fully recorded the variants in the Junagadh MS. I have also recorded the additional verses found in it (See Khilas II. 13; IV. 2). In the Kashmir MS. the beginning portion in Khila III. 20 was broken to the extent of four syllables. This Khila has fortunately been found in the Junagadh MS. I, therefore, filled in the blank on its authority by inserting यच्च कृतं found in it and have shown it in brackets. Scheftelowitz has conjectured the lost text as that. In two Khilas, III. 19; IV. 12 the Junagadh MS. presented a better text than the Kashmir MS. to a great extent. I have, therefore, adopted it in preference to the Kashmir text.

Scheftelowitz has, in his edition, also used two MSS. (No. 30 and 155 Müncher Sanskrit MSS. = Haug’s Verzeichnis d. orient. Handschr. No. 34 & 318) which he designates as M, and M2 respectively. Both these contain portions of the fifth Adhyāya. MS. No. 30 contains Nivids, Praişas, Purorucas and Kuntāpādhyāya. No. 155 contains only thc Nivids and Praişas. Scheftelowitz has recorded all readings from these MSS. Since, however, most of these readings were found identical with those from other sources, I have not referred to them. In very rare cases where the MSS. gave peculiar readings, I have recorded them without mentioning the source.

Khila IV. 11 is the Sivasamkalpopanişad coming after RV. X. 166. In the Kashmir MS. it consists of 13 verses. The same hymn has 28 verses in a MS. of R̥gveda Saṁhitā, No. Add 5351 in the British Museum, London. All the 13 verses in the Kashmir Khila are covered by the Khila in the British Museum MS., though in a different order. The Khila with the same 28 verses with identical text is also found in another MS., No. 2131 in the India Office Library, London. Scheftelowitz has publislied the text of the 28-versed Khila along with notes and German translation. (1. ZDMG 1921, Band 75, pp. 201-12.) I have reproduced it in the notes to the shorter Khila in the Kashmir MS. (2. In verse 7a the Kashmir MS. has lost one letter; the British Museum Ms. supplied it and made up the word गुह्यं.) This larger version of Sivasamkalpopanişad is also found in the Junagadh MS. The Aundh edition also gives the same Khila of 26 verses, some of which are different from those in the British Museum MS. The order of verses in the Aundh edition as well as the variants in the same are also recorded in the notes to the same.

Scheftelowitz has also made a text-critical and historical study of Khilas V. 1-4 after the publication of his book (1. Zeitschrift für Indologie und Iranistik, Band I, 1922, pp. 50–68.) and has recorded the variants found by him in certain MSS. (2. The Samjñana hymn is found in the following MSS.:-(1) Bodleian MS, Wilson 429-32, (2) Bod. Wilson 445, (3) British Museum MS. No. 5351, (4) India Office MS. No. 2131. The last three MSS, also contain the Nairhastya hymn and the last two contain the Pràdhvarānām. The Mahănāmnyah are contained in (1) British Museum No. 5356d Fol, 106B, (2) Br. Museum No. Add. 5351, (3) India Office No. 2131.)

Out of these, only those readings have been noted by me which were different from those already recorded, without mentioning the source. For full particulars the reader may refer to the original articles. The points of historical importance contained in them are added as English notes to the Khilas concerned.

shrIsUkta

Khila II. 6 is the well known Śrīsükta. This hymn addressed to the Goddess Sri is often read and recited in India from old times, as is clear from a large number of MSS. of the same found in all parts of India, and a number of commentaries written on it. It is also quoted in later Vedic literature as well as in a number of books dealing with religious rites. After the publication of the Kashmir Khila edition, Scheftelowitz collated six MSS. of Srīsākta in the British Museum, India Office Library, London, and the Bodleian Library, Oxford and published the critical apparatus along with an exegetical study of the same. Since the additional variants recorded there were found of no special significance, I have not recorded them here. (3. For the critical apparatus as well as for the study of the character of the deity and a German translation of the hyma, see ZDMG. 1921, Band 75, pp. 37–50.)

In this connection I may add that I have collated no less than 44 MSS. of Srīsūkta from different libraries like the Tanjore Maharaja Sarfoji’s Sarasvati Mahal Library, Tanjore, Government Oriental Manuscripts Library, Tirupati, Theosophical Society’s Library, Adyar and other sources. Since, however, they did not disclose different readings of any significance from those already recorded from other sources, I did not think it proper to make use of the material in the notes which have already exceeded their limit. A few independent readings found in them are, however, recorded without mentioning the source.

I must add that I also collated 8 MSS. of the Rātrīsūkta (Khila IV. 2) with accents and a MS.S containing four other Khilas also with accents from the Tanjore Library. (4. Nos. 329–36, T. M. S. S. M. Library, Tanjore. 5. No. 335. 6. Khila II. 1 supple. ; II. 14 with supple.; IV, 7 supple.; V. 1.5; Comp. Catalogue of T. M.S, S. M. Library, Vol. I. p. 201.) Since the MSS. did not exhibit any different readings from those already recorded, I thought it unnecessary to add their variants.

The two MSS. of Āśvalāyana Mantra Saṁhitā in the India Office Library ( Nos. 378 [1970] and 379 [781]) have recently been described (7. Annals of the Bhandarkar Oriental Research Institute, Poona, 1938-39, Vol. XX, Parts III-IV, pp. 249-61.) in detail by Dr. V. M. Apte, according to whom the work “gives all the R̥gveda Mantras cited by their Pratīkas in the Gșhya and Srauta sūtras of Ašvalāyana …… An interesting aspect of this work is that its versions of certain hymns are different in many respects from those in the editions of Müller and Aufrecht or in ‘Die Apokryphen des R̥gveda’ by Dr. Scheftelowitz.” I have recorded in the notes to the Khilas concerned the information regarding the different versions or other peculiarities contained in this work with regard to the Khilas. I have designated the MSS. Nos. 378 and 379 by आश्वमंसं1 and आश्वमंसं2 respectively. Where the text of both is identical, I simply mention आश्वमंसं.

Calcutta manuscripts

There are, in the library of the Royal Asiatic Society of Bengal two MSS. of R̥gveda (Saṁhitā, No. 120 (6607], Pada text No. 140 (6608]) which deserve attention. (1. Catalogue by MM. Haraprasad Sastri, Vol. II, Vedic MSS., Calcutta 1923.) I tried to procure these MSS. for consultation but could not succeed, since they could not be lent out, being sent out of Calcutta for safety in war period. I can simply record the peculiarities of these MSS. with regard to the Khilas on the basis of the information about then *supplied by the compiler of the Catalogue.

MS. No. 120 [6607] contains 12 batches of leaves. Batch VIII contains portion of Astaka VII probably written in the 18th century. It contains the Khila पावमानीः स्वस्त्ययनीः (III. 10 with supple.) conforming to its extent in Max Müller’s edition. Batch XII comprising Aştaka VIII is probably written in the 15th century. It contains the Khila अविधवा भव वर्षाणि (III. 17 with supple.). It also probably contains the Khila येनेदं भूतं (IV. 11 with supple.) with 28 verses as given in the British Museum MS. No. Add 5351 already mentioned, since the Catalogue quotes the verse गोबिर् जुष्टो धनेन which is the 19th verse in the abovementioned MS., and since the Khila is extended to 5 Vargas as stated in the Catalogue. Then again, it has the following Khilas : (1) संज्ञानम् उशना, (2) नैर्हस्त्यं सेनादरणं, (3) प्राध्वराणां पते and (4) विदा मघवन् विदा,, namely Khilas V. 1-4. Here the MS. ends. MS. No. 140 (6608] comprises the Padapātha of Aștaka VI written in Samwat 1653. The compiler says that the MS. “seems to belong to a recension of the R̥gveda different from that published by Max Müller. For instance, in VI. 4 there is one hymn more in Max Müller’s edition than in the present MS." The compiler probably meant some, perhaps the 10th, Vālakhilya hymn.

The 11 Vālakhilya hymns occur in the Kashmir collection (see Khila I. 4; 6 ; III. 1-8; 18). They are, therefore, reproduced in the proper place even though they were printed in Mandala VIII (Vol. III) in accordance with the usual practice. The Kashmir MS. exhibits a number of variants in their text. (2. For the value of these variants cf, Scheftelowitz pp. 35-45 and Oldenberg, Noten, II. pp. 116 ff.)

There is a commentary of Vidyaranya, among a few others, on the first 15 verses of Srisūkta. Max Müller has published the same along with the Srīsūkta in his R̥gveda edition. I have collated over a dozen MSS. of Srisūktabhāşya of Vidyāraṇya, as a result of which an improved text of the same is given in the footnotes to the Śrīsukta. I did not think it essential to record the different readings in the bhāşya, since they did not exhibit any material difference in its meaning.

Editorial Work

Scheftelowitz has based his edition of the Khilas on the Kashmir MS. The MS. has, however, certain incorrect readings which he has had to reject in preference to those either found in parallel passages in the Vedic literature or conjectured by him. In such cases, he has noted the incorrect readings in the MS. He has also consulted other editions of Khilas and has recorded their different readings. He has, moreover, made a comparative study of the Khilas with the help of the whole Vedic literature and has noted parallel constructions. While giving notes to the Khilas I have, to a certain extent, made use of the material so laboriously and skilfully collected by him. I have, however, verified each and every parallel quotation and, as a result of that, have made alterations wherever necessary. For this purpose, I have used new editions of Sanskrit works that could be available e.g. the Paippalāda Atharvaveda edition of Dr. Raghu Vira, Lahore, and the critical edition of the Mahābhārata being published by the Bhandarkar Oriental Research Institute, Poona. The verification of all parallel passages proved advantageous in that certain inaccuracies, more or less important, that had occurred as a result either of the printing mistakes or of oversight, could be removed. I have also made certain additions on the basis of the additional material already described. Certain additional verses and supplementary Khilas not recorded by Scheftelowitz have been noted by me.

In a good many Khilas the Kashmir MS. gave broken text and Scheftelowitz bas restored it in many places on the authority of parallel Vedic passages and has put it into brackets. I have adopted it as given by him, but have dropped the brackets. There still remained certain passages where the text in the MS. was broken. In respect of some such blanks, the Aundh edition proved useful. It contains some Khilas which are not found in Max Müller, Aufrecht and also in the Bombay edition and are evidently based on certain MSS. unknown to us. I have, therefore, adopted its text in such places and have shown it into brackets.

    1. R. A. S. Bengal No. G 5831 ; B. O. R. I, Poona No. 3 (1872-3); Panjab Univ. Lib. No, 281 ; T. M. S. S. M. Tanjore No. 652-3; G.O. MSS. Lib.. Tirupati No. D 23, D 24, D 17501, R 1979; Adyar Lib. No. IX, B 36, XXXV B9, XIX C 54, XXX F1, XXII R 6. XXII H 16, XXIX F 6, XXIX C 35.
    1. Cf. Khila II. 1, II, 6 (2nd supple.), II. 13, IV. 2.
    1. Comp. Apokr. des RV. Khila I. 3, 1-2, III. 1. 8-10, III. 10. 3-6; III. 11; III. 22. 5-10; IV. 5. 19-21; IV. 6.1-3; IV. 7,2; IV. 11. 9; V. 1. 5.
    1. See Note 1 p. 892. The different readings in all Khilas in the Aundh edition are not necessarily noted. Reference is made to addition verses, changed order of verses and important variants in certain Khila s.
    1. Comp. Khila IV.5. 7-9; 11 ; 12; IV. 11.9.

As already stated, I have filled in the broken word in the beginning of Khila III. 20 on the authority of the Junagadh MS. There still remain a number of lacunce in the Kashmir MS, which it is impossible to fill in for want of MSS. or other authentic material.

    1. Comp. Khila III. 15, 14; 22-27 ; 29; III. 20; IV. 9.2; 3.

As for the text of the Khilas, I have generally followed Scheftclowitz, though in doubtful cases I have referred to the original MS. Also, In certain cases, he has retained incorrect readings, but has suggested correct readings in the Notes. I have adopted some of such suggested readings in the text especially in the Sauparna hymns. In such cases where he has adopted a reading different from the Kashmir MS., I have not recorded the incorrect reading in it. I have also not referred to his notes in connection with the interpretation of the Khilas, determination of their age, emendations regarding the accents and other things not directly connected with the constitution of the text. In a few cases I have adopted readings different from Scheftelowitz, not conjecturally but on the authority of some parallel quotations or MS. evidence. In three cases I have adopted readings suggested by A. B. Keith in his reviews of Scheftelowitz’s work since the printed text seemed defective.

    1. Comp. Khila I. 2.3; 6; 7; 13; 1. 3,5; 1. 4.8; 1, 5.10; I.9.2; I. 12. 4.
    1. Comp. Khila II. 13,5 ; III. 3. 1 ; III. 13; III. 18, 1; III. 19; IV. 12; V. 3.4; V. 4.5; V. 22. 10 etc.
    1. Comp. Khila I. 2.4b; III. 15. 206; III. 16. 5b. 5. JRAS 1907, pp. 224-29,

I have given all notes in Sanskrit in accordance with our practice in all the volumes of the R̥gveda edition. In order to attain brevity, I have used certain signs the denotations of which are given along with the list of abbreviations of Sanskrit works. Notes or references in connection with works in other languages are given in English. Abbreviations in English, being common, are not given in the list. The Sanskrit notes are given in the following order:

  • (1) The Khilānukramanī to each Adhyāya, given at the beginning of the same, with variants, if any. (6. Macdonell in his Bfhaddevatā Part II has sometimes quoted the Khilānukramani which differs from Scheftelowitz’s text in some cases. Some of his readings have been accepted by me.)
  • (2) Place in the R̥gveda Saṁhitā where a Khila is to be inserted. Difference of opinion, if any, is recorded.
  • (3) Names of works (in abbreviated form) where the Khila occurs fully or in part.
  • (4) General notes as to the occurrence of the Khila in particular works, and different order of verses, if any.
  • (5) Variants in the order of verses. In each case the adopted text comes first in bold type with its authority and the variants with their sources follow. When the variant has very few sources, the authority of the adopted text is not denoted, meaning that it has got substantial support.
  • (6) Supplementary verses occurring in certain editions and other works.
  • (7) Variants in the supplementary verses.

Scheftelowitz has, in his work as well as in his articles written after the publication of the same, discussed many problems concerning the Khilas. I think it advisable to record his main points with regard to the antiquity of the Khilas and their relation to the R̥gveda Saṁhitā. Certain views expressed by him are not likely to be universally accepted, as is evident from the review of the work by A. B. Keith.

Antiquity of Khilas and their Relation to R̥gveda Samhita

The term Khila first occurs in the Anuvākānukramaṇi and Arşānukramaṇi attributed to Saunaka. Khilas are those hymns or verses which are not included in the Saṁhitā and Pada text of the R̥gveda in the Sākala recension and as such are excluded from the enumeration in Sarvānukramani, Anuvākānukramaṇi etc., while their position within or without the recognised hymns of the R̥gveda is traditionally fixed.

The Khilas belong to the Vedic period. Khila verses coming in other Vedas and Brāhmaṇas are the oldest proof for their original form and ritual application. In Brāhmaṇas themselves there are Khilas (e.g. Nakulasūkta, Kuntapa, Vālakhilya) used in ritual where they do not suit according to their contents and their original purpose. This shows the high value and sanctity that the Khilas received already before the Brāhmaṇa period.

Individual Khilas do not belong to the same period. Many reach back to the oldest rgveda period (e.g. Purorucas, Nivids, Praişas). Many belong to the end of the R̥gveda period (e.g. Mahānāmni, Vâlakhilya, Kuntāpa). Many belong to the oldest Yajurveda period ; some are later Vedic texts and presuppose knowledge of many Yajurveda hymns (e.g. Subhesaja hymn). For the determination of the age of individual Khilas each one must be examined separately. The general conclusion is that all Khilas reckoned in the Khilānukramanī belonged to the Vedic period and that most of them belonged to that period in which the Yajus, Sāma and Atharva Saṁhitās were compiled, and stood in the same condition in which they are preserved today. Very few Khilas belong to the later Brāhmaṇa period. A good many must have been composed before the peculiarities of the liturgical practice of Brāhmaṇa period were established, since most of the Khilas were applied in ritual without standing in close relation to particular rites in which they are used. On the other hand, the later Khilas which originated in the Yajurveda or Brāhmaṇa period are usually composed in connection with the ritual which gradually arose in that period.

  1. With regard to the date of the Khilas, A. B. Keith bolds that “the language, style and contents of the vast bulk of the Khilas undoubtedly belong to the latest period of the R̥gveda. It is quite possible that some verses now preserved as Khilas were parts of the Mandukeya sākhã, but the probability is not great. The really significant fact is that Sākalya did not deal with Khila’s in the Padapātha………. It is most probable therefore that the Khilas are somewhat later in date than the R̥gveda Saṁhitā. It is undeniably the case with a good deal of the matter and in no case is it impossible.” JRAS 1907, pp. 224-29,

Thus, for example, the Medhāsūkta stands in close relation to the Upanayana ceremony that comes for the first time in AV. XI. 5, the Subhesajasūkta is presupposed as well known in Gopatha Br. V. 23 and the Srīsākta stands in relation to the worship of Sri and Lakşmi coming in Yajurveda period. The eleven Suparna hymns occurring after RV. I. 73 are, according to the Khilānukramani, composed by the authors of the Suparna race. The text of these hymns was unknown before, even though certain sources in the old works had denoted their existence.1 The Khilānukramaṇī describes these hymns more closely ; it gives their Rşis. It also says that the authors whose race is not given here, belong to the Suparṇa race.2

    1. R̥gvi. I 108 ; Brhad. III. 119; Aśva. GF. III. 12. 14 ; ABr. VIII, 10.4; V[ 25.7 with Sāyana.
    1. For the place of Suparna hymns.in R̥gveda see Scheftelowitz, ZD.MG 74, pp. 192-203.

Sākalya, who redacted the R̥gveda Saṁhitā lived, as Geldner has shown, in the later Vājasaneya period; he was a contemporary of Aruni mentioned in many Brāhmaṇas, as well as of Yājñavalkya who flourished a little later. Thus the sākala sākha is as old as the Vājasaneyi Samhita. Sākalya can therefore scarcely be placed, following Oldenberg, at the end of the Brāhmaṇa period.

    1. A. B. Koith is not inclined to accept Geldner’s view. He says, “Oldenberg’s arguments (Hymnen des R̥gveda, Baad 1, pp. 384 ff.) for the priority of the Brāhmanas to the adoption of the rules of Sandhi and the still later Padapatha are not adequately met.” JRAS 1907, p. 226.

Now most of the Khila texts existed long before the time of Śākalya and were enveloped by the halo of sanctity. Why then did Sākalya not incorporate either the new or the old apocryphal texts in his codex? And why did he not compose a Padapātha for them? He has treated them in the same way as the following verses in R̥gveda : VII. 59.12 ; X. 20.1; 121, 10; 190. But while these Khila verses, not analysed in the Padapātha, were incorporated into the R̥gveda Saṁhitā, were considered as recognised texts, were later also provided with a Padapātha and were accordingly enumerated in all Anukramanīs, why did our Khilas not obtain the same treatment as those R̥gveda verses not looked upon as of equal authority by Sākalya ? In fact many Khila verses in our Khila collection are really older than those redacted by Sakalya. The only explanation seems to be that the Khila texts belonged to a recension different from that redacted by him. The fact that many old apocryphal verses did not originate independently, but were very closely related by contents to those R̥gveda hymns to which they are annexed, makes it probable that originally many R̥gveda hymns had been existing in several recensions, just as a hymn of the Paippalāda sākhā of Atharvaveda is distinguished from the corresponding hymn of the Saunakīya Atharvaveda not only by its greater extent but also by a different order of verses and by different readings.

The case of Purorucas, Nivids and Praişas in the fifth Adhyâya of the Khila collection, to which no particular place is assigned in the R̥gveda, is rather different. They belong to the older R̥gveda period.

    1. Comp. AV. XI. 7. 19; VS. XIX 25; TS, II 5. 9; ABt. II 3,4; SB. I. 4. 2.

The Nivids form the oldest prose preserved from the R̥gvedic time. They are certainly older than certain R̥gveda hymns. Many R̥gveda verses appear to be poetical compositions from individual Nivids. The word Nivid comes already in the R̥gveda as the term for these prose texts. The Praişas were considered as an appendix to the R̥gveda in very old time.3 The Purorucas also belong to the same period as that of Nivids and Praişas.4 According to Scheftelowitz, the character of Nivids, Purorucas and Praişas lends support to Hillebrandt’s theory that along with the tradition which orally transmitted the treasure of hymns, a tradition relating to the ritual technique developed independently and the majority of Vedic hymns had a place in the ritual. The Saṁhitā was certainly not edited on the basis of the hymn material used in the ritual on account of the many very small ritual ingredients ; on the other hand, the hymns used in the ritual did not stand in need of a redaction in the Samhita in order to be inserted in the ritual. Several old verses which belong to certain R̥gveda hymns in accordance with the ritual tradition are not incorporated into the R̥gveda Saṁhitā, but have come down as Khilāni. That all Khilas which have not found admission to R̥gveda must have belonged to the so-called ritual tradition, cannot be accepted, because certain very old Khilas (i. e. Suparṇa, Vālakhilya, Kuntāpa, Nakula, Mahānāmni) originally stand in no relation whatsoever to the ritual in view of their contents, in which they are used in the Brāhmaṇas, Srautasūtras and Gșhyasūtras, but are used in the ritual only secondarily. The Khilas composed in later Vedic period were composed from the first in connection with certain newly accepted ceremonies.

    1. For the antiquity of Nivids and Praişas see Scheftelowitz, ZDMG Band 73, 1919, pp. 30-51.
    1. See Rkprāti, 1. 57; Bghad. VIII. 103 ; Nir. VIII, 22.
    1. Regarding the antiquity of Purorucas, see Scheftelowitz, ZDMG, Band 74, 1920, pp. 204-7.
    1. H. Oldenberg has raised an objection to this theory of Hillebrandt (Hymnen des R̥gveda I. pp. 519 ff). A. B. Keith also does not accept the theory; cf. JRAS 1907, p. 226, See also W.D, Whitney, Atharvaveda, p. 848.

The examination of the Khilas leads to the conclusion that they were really admitted to some lost śākhā. This is proved by our information of the Bāşkala śākhā. The Anuvākānukramaṇi (verses 21, 36) provides us with some peculiarities of the Başkala śākhā. Thus in it the Kutsa hymns (RV. I. 94–115) stood after the Parucchepa hymns (I 127-139), then it enumerated eight hymns more than the Sākala recension of 1017 hymns. The modern commentator of Caranavyūha, mentions these eight hymns: The Başkala sākhā placed the first two Vālakhilya hymns at the very place where we read them today. The next five Vālakhilyas followed R̥V. VIII. 94. At the end of the whole Saṁhitā, after X 191, followed the Sarjñana hymn of 15 verses in the Başkala śākhã. That the Samjñāna hymn of the Başkala sākhā originally consisted of 15 verses is impossible, for then the Başkala śākha should have 10 more hymns and not 8, since the so-called 15-versed Samjñāna hymn is really divided into three separate hymns. The confusion seems to have arisen from the fact that the same verse TacchaMyorāvrnimahe forms the last verse of the first as well as the third hymn.

The 8 (10) regular hymns of the Başkala recension were considered in the Sākala sākhā as Khilāni. Now a good number of Khilas are as old as the Vālakhilyas and the Samjñāna hymn, and many of them are still older and are already used in the ritual of the Brāhmaṇas belonging to the R̥gveda just like the R̥gveda hymns. The fact that the Brāhmaṇas and Srauta and Gșhya sūtras treat the verses called Khilāni at present just like Rcah and quote them by merely the beginning words just as in the case of Rcah, leads to the conclusion that they had been standing in a collection related to the R̥gveda in their time. The wholly extinct Māṇdūkeya sūklia which is mentioned in Rkprāti. III. 14 and Aitareyāranyaka could have admitted all old Khilas as equally authentic texts. According to the Purāṇic tradition (- 1. Devi Purāṇa says :- शाखाश् च त्रिविधा भूप शाकला बाष्कमाण्डकाः। Comp. Weber, Ind. Stu. III. p. 253.), there were three śākhās, sākala, Başkala and Mândukcya.

    1. According to the Caranavyuha there were five Sākhās of R̥gveda. Of these, according to Oldenberg, the Asvalāyana and Sankhāyana were looked upon as independent Sakhis only in a restricted sense. So there remained only three, which number is also referred to in the Purāṇic tradition as stated above. Of these, again, the Mandūkeya remning out of consideration, because the name of Māṇdūkeya though mentioned in the Aranyaka and Prātiśākhya is not at all concerned with the sākhā of the R̥gveda. The knowledge of this Māṇdūkeya śākhā was lost even in early period. So, according to Oldenborg, the only sakha that requires a thorough examination besides the Sākala, is the Başkala Sakhā (Hymnen I. pp. 490-91).

According to the commentary on Caranavyűha, the Asvalāyana sākhā was formed simply by adding eleven Vālakhilya hymns. The difference in the Sankhāyana was that the 10th Valakhilya hymn (RV. VIII. 58) was absent, the first two verses of which were inserted as a Khila after RV. X. 88. 18. This Khila is found in the Kashmir MS. It is to be noted that the two MSS. of the Sankhāyana śākhā of R̥gveda Samhită procured by Pandit Satavalekar omit the 10th Vâlakhilya and give its two verses as a Khila after RV. X. 88. 18. From the statement in the Caranavyūhabhāşya regarding the Sānkhāyana sākhã we can be allowed to surmise that that sākhã contained certain Khilas even though they were not enumerated as part of the Samhita. Oldenberg says that the quotation in Sankhāyana of the Sandhi termination Upadruta which is peculiar to the Başkalas and the allusion in Sankh. Gr. IV. 5.9 of Tacchanyor as the last verse of R̥gveda, suggest the probability of Sankhayana being a follower of the Başkala sākha (Hymnen I, p. 517).

The Khilas admitted by the other recensions existed at the time of Sākalya who fixed the present order of hymns. He seems to have excluded them only because different hymns and verses had not come down in a unanimous tradition. That many hymns in the Khila collection did not hold an undisputed position in the R̥gveda Saṁhitā is evident from the fact that certain Khilas (e. g. II. 15, III. 17, IV. 14) are fully quoted by Āśvalāyana and Sankhāyana, while usually they cite merely the beginning words of the R̥gveda hymns and most of the Khilas.

Scheftelowitz follows the arguments of Benfey (Samaveda sxviii) and Weber and holds that originally Sāmaveda arose from the same material from which the R̥gveda was compiled. This is supported by the fact that in Sâmaveda there are 71 verses which are not in the R̥gveda, some of which are preserved as Khilas to the R̥gveda. It is also argued that Sāmaveda contains the R̥gvedic verses in an older form since the variants in the Sāmaveda are generally in a more abstruse and incomprehensible form than those found in R̥gveda in their places. Then following Hillebrandt’ he argues that many hymns composed in the R̥gvedic period are preserved in Atharvaveda or Yajurveda and they cannot be considered as later because they are found in Atharvaveda or Yajurveda. It is, therefore, quite natural that several Khilas are found in them. In this connection it is to be noted that Benfey’s arguments regarding the older tradition of Sāmaveda are not proved. Theodor Aufrecht has shown that the divergent readings of the Sāmaveda are due only to arbitrary, intentional or accidental alterations.2

    1. GGA 1889. p. 406.
    1. Cf. Winternitz, History of Indian Literature, Vol. I, p. 164; A. B. Keith, JRAS 1907, p. 227.

In the codification of R̥gveda, Sākalya seems to have recognised only those hymns which had already become general property and were recognised by all as traditional national hymns ; on the other hand he seems to have considered those hymns as doubtful Rcas (Khilani) which lead an isolated life in the tradition of individual families. And in those cases in which he could not decide whether a particular verse had really come down from old with equal authority, he appears to have admitted it in his Saṁhitā, but did not analyse it in the Padapātha, like RV. VII. 59. 12; X. 20.1 ; 120. 10; 190. On the other hand, another recension (like the Başkala sākhā) recognised several hymns from the doubtful material as authorised text and incorporated them into the Saṁhitā, and again another sākhā (like the Māṇdūkeya) might have accepted the whole old Khila portion as authorised text. However, they were considered as sacred texts by the Sākala sākhā even though it denied them equal rights with the R̥gveda hymns, as is evident from its descendants in the Srauta and Gșhya sūtras as well as Saunaka. The Başkala šākhā has also preserved the texts not recognised by it, for Bșhad. VIII. 85 states that the Başkala śākhā considered the verb Adadhe in the second verse of Nejameşa hyma as third person singular perfect. This Nejameşa hymn is found in the Khila collection. Thus the Khilas were treated by the Başkalas with the same accuracy as the R̥gveda hymns. This alone can explain the fact that these apocryphal verses and hymns were never denoted by the term Khila in the oldest time but were always regarded as Rcas. Although at the time of the compilation of the Aranyakas all the Khilas mentioned in the Khilānukramaṇi already existed, and although many citations from them are found in the Aranyakas, the term Khilâni did not exist in that period, for in TĀ II. 9. 10 where all sacred texts are enumerated, Khilāni are not mentioned.’

    1. In this connection Keith observes that " it is not to be expected that so paltry a species of literature as the Khilas would find a place in such enumerations of the great literary genera.” JRAS 1907, p. 227.

Looking to the old works relating to the study of Vedic literature we find that the Khilas were looked upon by them as Rcas and the term Khila was unknown to them. Yaska in his Nirukta cites several Khila verses and denotes them as Nigama.2

    1. Comp. A, B. Keith, ibid. p. 226 ff. ; " Examination will show that all the cited texts occur in other Vedic texts, and in any case such citations could not prove that he considered the passages as belonging to the R̥gveda Saṁhitā."

In Saunaka’s time the term Khila did not exist; Khilas were considered by him as Rcas that really belonged to the R̥gveda but were not counted in Anuvākas. He, therefore, explains Khilas in his Prātiśākhya.

    1. Comp. A, B. Keith, ibid." In the case of Saunaka that he did not know the Khilas as such is still less convincing. The fact that the Khilas are included in Prātisakhya proves nothing except what is admitted, viz., that the Khilas existed in connection with the R̥gveda at the time of the composition of the Prātiśākhya which, it may be added, is quite possibly later as we have it than Saunaka."

Brhaddevatā and R̥gvidhāna also do not know the term Khila and understand by R̥k verses also the so-called Khilas. Both these works are not composed by Saunaka himself but by his followers. Among the R̥gvedic deities BỊhaddevatā mentions also those which relate only to the apocryphal hymns.

    1. Thus Lākşā (1. 129, II. 84), Šri (II. 84), Medha (II. 84), Nişad and Upanişad: (II. 82); comp. Khila IV, 7, II. 6, IV. 8. I, 3.

A good many Khilas from the Khila collection are treated in Brhaddevatā. There are, besides, certain citations in it which are found neither in the R̥gveda nor in the Khila collection. (5. See Byhad. VII. 104, I. 48, IV. 87-8, V, 157)

The R̥gvidhāna deals with 23 Khilas in this collection. (6. Sauparna hymns (I. 2-12), Srīsūkta (II.6), Agniretu (II, 11), Samvatih (II. 13), Uttudainam (III. 16), Yan Kalpayanti (IV.5), Āyuşyam (IV,6), Medhan Mahyam (IV.8), Yenedam (IV. 11), Nejameşa (IV, 13), Samjñānam (V. 1), Nairhastyam (V.2), Mahåndmnyaḥ (V. 4). )

From the absence of many Khilas in R̥gvidhāna or in Bịhaddevatā one cannot conclude that at the time of the composition of both these works these excluded Khilas were not recognised ; for many R̥gveda hymns themselves are not quoted in them. While in Rkprātiśākhya and in Bșhaddevatā the Vālakhilyas and the Subheşaja hymn are mentioned, they are not treated in the R̥gvidhana ; on the other hand the hymn Samvatīḥ (II. 13) is not mentioned in Brhaddevatā while it is treated in R̥gvidhāna. These Khilas were, however, not counted in the enumeration of R̥gveda hymns.

    1. “The evidence of the Brhaddevatá shows that Saunaka did not include these hymns in his enumeration of the hymns of the R̥gveda. Why he did not do so is of course open to dispute, but it is perfectly fair to argue that he may have done so because they were in his day regarded as not R̥gvedic in the proper sense and the evidence given above is in favour of his having used the term Khila of them. That this expression does not occur in the Bfhaddevata or the R̥gvidhāna is no more surprising than it does not occur in tho Prátisakhya."-A, B. Keith, ibid. pp. 227 8.

The Anuvākānukramaṇī and Arşānukramaṇi are said to have been composed by Saunaka. Since, however, they expressly use the term Khila,” they cannot be attributed to him.

    1. Comp. Anuvākānu. Verses 7, 17, 36, 39.

Arşānukramanī is much later than Anurākānukramaṇi; the composer of the former work holds the Khila collection as a unitary work composed by Gautama Vāmadeva. ( - 2. Arşānu. X. 102. For different views regarding the date and authorship of Arşānu. kramani, see Macdonell, Bșhaddevată, Part I, Preface, p. xxiii ; Festgruss an R, von Roth, p. 112 ; A. B. Keith, JRAS 1907, pp. 226 ff.)

Āśvalāyana and Sankhāyana do not know the term Khila. The sanctity of Khilas was much degraded in the period of Dharmaśāstra. (- 3. Comp. Manu. III. 232.) Şadguruśişya, the commentator of Sarvānukramaṇā who flourished at the end of the 12th century A.D., had the same ignorance about the Khilas as at present. (- 4. Comp, Sarvãou, I. 99,) The ignorance regarding the Khilas gradually grew so far that several modern popular hymns with religious contents which originated in connection with popular cults (e.g. Serpent cult, Durgā-cult) began to be considered as Khilāni. These later hymns are partly added to some old Khilas without regard to their different contents, so that consequently several Khila texts were affected by modern supplements (e.g. II. 1; 6; 12; 14; III. 10; 12; 17; IV. 20). Only in a few quite modern Khila MSS. such later texts are given as independent texts. The term Parisişțâni is met with only in modern works like the Prayogaratna.

Acknowledgments

I have to perform the pleasant duty of acknowledging the help received by me in accomplishing this work. I have already stated that I have chiefly based the text of the Khilas on that in the Kashmir MS. published by Dr. Scheftelowitz in his “Apokryphen des R̥gveda.” It is, therefore, my first duty to acknowledge my indebtedness to him. I began the study of the Khilas three years ago when the dark clouds of war had suffocated the atmosphere of the whole world. Copies of Scheftelowitz’s work were not easily available, and it was with the utmost efforts that I could procure one for my purpose. It is a matter of relief that the war has come to an end before the publication of this work. I must also acknowledge the help I obtained from other editions of Khilas and other Sanskrit works, and I express my thanks to their editors. To the various MSS, libraries I am indebted for allowing me to use the valuable MSS. of Khilas in their possession. I have especially to thank Pandit Satavalekar of Aundh (Dist. Satara) for kindly keeping at my disposal his two MSS. of Sãokhāyana sākhā of R̥gveda. I have also derived help from the writings of Oldenberg, Hillebrandt, Macdonell, Keith, and others to whom my thanks are due. I must also acknowledge the help I received from certain scholars who kindly procured for me certain valuable and rare printed books. Last, but not least, I must also thank those scholars who gave me their valuable advice with regard to the preparation of this work.

C. G. KASHIKAR