1. Ed. V. Raghavan, Madras 1949, pp. 251-3. 2. The approximate number entered is as followa: MSS without commentaries about 250 and MSS with commentaries about 200; the latter item is distributed roughly thus: Anonymous glosses or commentaries 40, Vema 88, Ravicahdra 25, Arjuna-varman 20, Koka 4,Caturbhuja 4, Rudrama 6, Surya-dasa 3, other commentaries of known authorship 10. The numbers are not only approximate, but also sometimes illusory because of wrong or uncertain entries in the original catalogues, while some mentioned in Reports of Search of MSS are already included in particular collections of MSS libraries, or are untraceable. 3. And in most Indian libraries loan, reliable trancripts, photostat or rotograph copies are either difficult to abtain or not available. 4. As the painstaking edition of D.D. Kosmabi (Bombay 1945) would show. 5. This is only a general statement which should not be taken too literally. In the BORI (poona)ed. of the Mahabharata, where a very large variety of MSS of different recensions and versions has to be dealt with, this principle is generally follwed; but the question would depend on the nature of the text and kind of MSS available. 6. None of the three copies of this edition that we have consulted contains any title-page, giving the date and place of publication, although the pagination, type and general format of all these copies are idential. The copy in our own possession bears on the fly-leaf the handwritten description : "E(ast) I(ndia)College/No. 30/Library"; pp.1-117; followed by the Ghatakpara Kavya with an anonymous commentary, pp. 1-15. The copy, now existing in the library of the Presidency College, Calcutta, appearsto have belonged originally to the College of Fort William, Calcutta, as the rubber-stamp of this College indicates. It is a badly wormeaten and damaged copy (also with no title-page);the final forme containing the Ghatakarpara Kavya is missing; otherwise the copy shows no difference. Another copy received on loan from the National Library of India (Calcutta), also appears to have belonged originally to the College of Fort William. It is better Preserved but pp. 55-6 and 93-4 appear to have been misplaced in binding respectively with pp. 63-4 and 95-6;no page, however, is lost, and the Ghatakarpara Kavya is found in its place; pp. 1-117, 1-15. There is no title-page, but on the fly-leaf there is an old library note: "Amarusatakam by Amaru. 1808. Calcutta". The total number of stanzas in all these copies is 100. These copies appear to be the same as that used by Simon in his edition, and described as: "Text des Amaru mit dem Commentar des Ravicandra. 100 verse. Zusammen mit dem Ghatakarpara gedruckt. Calcutta 1808." It is not clear if Simon's copy contained any title-page. The title-page is missing in the India office copy [see Catalogue of (Printed) Sanskrit Books, ii,pt.1,revised ed.,London 1938]; but the title of the work is said to have been supplied "from the colophon", although it is noted from what source the date and place of publication (viz. Calcutta 1808) is supplied in enclosing brackets. So also in the British Museum Catalogue of Sanskrit (Printed) Books (Haas). J. Eggeling in his Catalogue of Sanskrit Manuscripts in the British Museum (p.100) also note that the work was printed at Calcutta in 1808. So far as we have been able to ascertain, this bibliographical information was first supplied by Adelung (An Historical Sketch of Sanskrit Literature, Eng.trs., Oxford 1832, p.35) where the date given is 1818. (obviously a misprint for 1808), and repeated by J. Gildemeister in his Bibliothecae Sanskrita (Bonn 1847), no. 162 (264), p. 73, where the date and place given are Calcutta 1808. Ravicandra's commentary appears to have been printed also in an edition of the text published from Benares; but we have not been able to trace the edition. 7. This EDITION GIVES NO INFORMATION ABOUT THE MANUSCRIPT-MATERIAL UTILISED. iT IS CURIOUS THAT THE TEXT OF THIS EDITION IS REPRODUCED VERBATIM IN THE TWO modern anthologies of Haeberlin and Jivananda published from Calcutta in 1847 and 1888 respectively (see below). 8. Das Amarucataka in seinen Recensionen dargestellt, mit einer Einleitung und Anzugen aus den Commentatoren versehen von Richard Simon. C.F. Haesler: Kiel 1893. The text is given in Roman transliteration. This edition is important not only because it presents Vema's South Indian text but also because Simon for the first time distinguishes the various recensions of the text. For the other recensions, from which additional stanzas and variant readings are noted , Simon utilised (i) three MSS of Ravicandra's commentary as well as the of this text, two MSS of Kokasambhava's commentary, and one Devanagari MS without any commentary. But Simon distinguishes a fourth recension, which he calls gemischte (mixed). For this he utilises one MS of Rudramadeva's commentary in Nevari characters, one each of Ravicandra's and Ramarudra's commentary in Bengali characters, as well as two Devanagari and four Bengali MSS without any commentary. 9. Printed at St. Thomas press, Cochin 1881 in Malayalam characters. No. MS of this commentary is noticed anywhere. 10. Our references are to third revised edition of 1916 by Durgaprased and Panasikar. This edition utlises (i) three MSS of Arjunavarman's commentary (two of which are dated Samvat1665 and 1695==ca. 1609 and 1639 A.D respectively). The first of these MSS appears to be the same as BORI MS No 324 of 1892-95. But the edition also makes use of (ii) two MSS of Vema-bhupala and (iii) one MS of Rudramadeva, from both of which additional stanzas are given. The MS of Rudrama, dated Samvat 1579 (==ca.1523 A.D), appears to be the same as the BORI MS No.457 of 1891-95. 11. The sense in which the term `recension' is used is explained below. 12. JAOS, vii, 25,33; JRASB v,p.378f. Cf. D C. Ganguly, History Of the Paramara Dynasty, Dacca University Dacca 1933, p,200f.--Madana Balasarasvati of Gauda was the author of the drama Parijata-manjari or Vijayasri, only two Acts of which have been preserved in a stone inscription at Dhara. It has been recovered and edited by E. Hultzsch, Leipzig and Bombay 1906. It was composed at about 1213 A.D. On this drama, see S. K. De, History of Kavya Literature, Calcutta University 1947, p. 472. 13. As he is described in the colophon to the Madras Oriental Library MS No. 11921 (p. 7979 of the Catalogue, xx). 14. See S. K.De, Sanskrit Poetics, i.p.314. 15. The three chief commentators appear to have thier own points of view. Arjunavarman interprets the text from the literary (Kavya-Alamkara) aspect, and appropriately names his commentary Rasika-samjivani, while Vema concentrates upon the erotic significance in consonance with the Rasa-sastra, and accordingly calls his commentary Srngara-dipika. Pischels suggestion that originally the Amaru-sataka was intended, like Rudra's srngara-tilaka (Preface to his ed., pp. 9-11), to illustrate the sentiment of love and types of heroine, etc., is ingenious but very unlikely. On the contrary, the Sataka seems to have been an independent work of great poetic beauty (see S. K. De, History of Kavya it from different points of view shews that there was no definite tradition of any particular object associated with the work. 16. The MS reads: khana-sri-ravicandra esa (instead of srila-sri-ravicandra esa of the printed text and most MSS). Sri Dinest Chandra Bhattacharya informs me that a MS in his possession has the same reading. 17. Some notable Bengali poets, patronised by the Muhammadan Court, appear to have had this title, e.g., Gunaraja Khan, author of the Bengali poem Sri-krsna-vijaya, etc.--Curiously enough, the MS in Mitra (Notices, ii, No.557) reads in the colophon : kaladhara-sena-krta amaru-sataka-tika samapta. This obviously led Bendall in op.cit.to mention the name as Jnanananda Kaladhara Sena. 18. As a matter of fact, citations and references to Rudra-bhatta's work occur throughout.-- The Rasa-Pradipa cited by him (on st. 39, 43) is not the work of the same name by Prabhakara Bhatta. 19. The name of the commentary is not Vidagha-cudamani as sometimes alleged. The concluding Arya verse, from which this erroneous impression arises, runs thus : amaruka-satakam idam ittham sva-buddhi-vihhavad rasabdhi-tattvajnah/ rudrama-deva-kumaro vidagha-cuda-manir vyavrnot// where the epithet vidagdha-cudamani applies to the author and does not signify the name of the commentary. Really there is no name; but from the verse the name would be simply Vivrti; or perhaps it is Avacuri (marginal gloss) given by the Florentine MS, as well as by one of the BORI MSS we have consulted. It consists of brief glosses on words and phrases, and is described as a Tippanika in the colophon to the British Museum MS. 20. yah-sahitya-sudhambhodhau nisnatah kokasambhavah/ tanute marukasyasau tikam nistankitasayam// 21. The name Rudrama is somewhat peculiar. Rudramma (or Rudramaba) is of course a South Indian name, having been borne by Rudramma Devi, daughter of Ganapati of the Kakatiya dynasty, who ruled over Eastern Deccan. But this female name perhaps has no connexion. 22. The British Museum MS No. 256 Or 3566 (100 stanzas; C.Bendall's Catalogue, London 1902,p. 100), utilised by Simon pp. 12-13 (==Simon's P); it is described in the Catalogue as a palm-leaf MS of "the 15th-16th century Nepali writing'. Only 6 MSS of this commentary are reported to exist. We have used two complete MSS belonging to Bori No. 457 of 1891-95 (dated Samvat 1579==ca.1523 A.D.) and No. 366 of 1877-91 (dated Samvat 1643==ca. 1587 A.D.). They are both in (Jaina) Devanagari characters. They contain, however, 114-115 st. as against 100 of the Br. Mus. MS, of which we posses a rotograph copy supplied by the courtesy of Dr. L.D. Barnett. Another MS is noticed by Aufrecht in his Florentine Sanskrit Manusripts, Leipzig 1892, No. 75, but this could not be obtained by us. There are two other incomplete MSS in the Bori which we have also cnsulted, but Simon did not have access to these four BORI MSS. One of these incomplete MSS (BORI No. 456 of 1891-95) bears the date 1440-41 A.D. 23. The question will be discussed below. By the term `recension' we mean generally a distinct cleavage in text-tradition which results in a particular arrangement and particular time and space. Within the recension it is possible further to distinguish `versions', given by smaller groups of MSS of a common type; but of course these are generally of a less probative value. A stray unsupported MS is of little importance for purposes of textual criticism. 24. J. Eggeling, Catalogue, vii. Londion 1904, p. 1521. 25. These two MSS (C1 and Cx) have almost the same order of stanzas, the total number in both being 106. But C1 expressly says santa-rasam atra vyakhyasyamah, while Cx has no such object .--The Vangiya Sahitya Parisad (Calcutta) MS (fol. 1-26) appears to be the same as C1, but it breaks off without colophon with the st. tanvangya guru-samnidhan (counted as 100) at fol. 26a, although it writes one Pada of the next st. bhru-bhedo gunitas ciram without any commentary on fol. 26b. Like the India Office MSS it is written in Bengali characters. We had a loan of this latter MS through the courtesy of Sri Dinesh Chandra Bhattacharya. 26. Notices, vii, Calcutta 1884, No. 2367, p. 135. Folio 1-28; no date, but appearance of the MS described as "fresh". Probably a modern copy. The India Office MS appears also to be a modern copy, not older than the 19th century. 27. The use of the strange phrase kastasrstya (as a synonyn of katham cit or katham api), familiar in the vocabulary of Bengal scholiast, occurs in its comment on st. 90 (ayate dayite, India Office MS, fol.60b.)! The phrase also occurs in Ravicandra's commentary (katham api kastasrstya) on st. 39. 28. The question of thier affiliation will be discussed below. 29. Only 4 MSS of his commentary are reported. We have used BORI MS No. 69 of 1883-84, in which the text does not accompant the commentary. It is perhaps the same MS as used by Simon (CK), p. 11. --Koka cites Rudra's Srngaratilaka (i.31)in his commenton st.7. By name he quotes, among other works or authors, Bharata-sutra (fol. 2a)--Kavya-prakasa (fol. 36). Haima-kosa (fol. 4a marg.), Visva, Amara and Halayudha (lexican fol. 14a). 30. In BORI MS No. 321 of 1884-87, which we have used. So far only 4 MSS of this commentary are known to exist. 31. Kampilye svardhuni-nira-pura-tira-tara-sthite/ cuturbhujena vasata bhava-cintamanih krtah// See Postscript below, p.48. 32. BORI MS No.365 of 1887-91, which bears the date in Samvat 1726==ca.1670 A.D. The colophon to this MS gives the name of the commentator as Misra-Caturbhuja, but this is not found in the other MS. The verse quoted above (f.n.) is missing in this MS,but it is given also by BORI MS No. 364 of 1887-91 (Gode, Catalogue, pp. 12, 25).--The definitions of Alamkaras in this commentary are drawn mostly from Mammata's Kavya-Prakasa. 33. The colophon to the BORI MS No320 of 1884-87 (fol. 62a) states : likhita ajameramadhye atmartham/samvat 1764 varse caitra vadi trtiya bhrgu-vasare// As no scribe's name is found, the phrase atmartham presumably refers to the author himself. But the Cimani-caritra of Nilakantha Kavi which follows (fol. 62a-71)in the same MS and is written by the same hand also hears a similar inscription with the words likhita atmartham only ! --A MS. of this commentary (fol.1-61), dated Saka 1642(==1720 A.D), is noticed by H.I.Poleman (Census of Indic Manuscripts in the United states and Canada, New Haven, Conn. 1938, p. 99, No. 2130)as existing in the Harvard University Library. It is a composite Ms copied by two different hands. 34. In the preface to his ed. of Amaru-sataka mentioned above. 35. BORI MS No.271 of 1884-6. 36. A commentary by Sesa Ramakrsna, which is said to have followed Vema-bhupala, is mentioned by Durgaprasad (op.cit.), but no MS of it is registered anywhere. We have not been able to obtaon the commentary of Vidyakara Misra mentioned in Jayaswal's Mithila Catalogue, ii, Patna 1933; while the commentary of Harihara Bhatta and the Amaru-darpana are known only by name from Buhler's Catalogue of Skt MSS in Private Libraries of Gujarat, Kathiawad, etc. Bombay 1871-73.--Weber's Berlin MS (Verzeichnis desSanskrit-Handschriften d. Koenigl. Bibl. i, No. 585) is a Devanagari exemplar of the text with no commentary, beloning apparently to the Western recesion (==Simon's B). 37. Having been praised by Anandavardhana as a well-known poet of eminence for the delectable and suggestive content of his erotic stanzas. 38. This is only a rough approximation. The inter-relation between MSS is much more complicated than this simple division into three recensions; and it would be possible to draw fine distinctions and group versions within recesions. Some of this Possiblity are indicated below in our general survey of the recensions. 39. It contains the commentary as far as the 25th stanza (only 10 folios). 40. The numbering indicates the serial number of the stanzas in the particular text. Except in special cases, only included (and not excluded) stanzas are noted with relevant references to the particular version or recension. Where there is no such indication, it should be understood that the particular version or recension, not mentioned, omits the particulr stanza.--Other abbreviations employed : comm.==commentary ;anon==anonymously; Simon==Simon's edition as above ; om.==omit; st==stanza; trsp==transposed. 41. The three references are to the numbering, in order, first of the Br. Mus. Ms and then of the two BORI MSS No. 366 mentioned above. Between the two BORI MSS the difference in the total number of stanzas is slight, but the Br. Mus. MS gives only 100stanzas. as against thier 114-15. See Appendix for a comparative table of the sequence and choice of stanzas in these MSS. 42. Although some defective or differently arranged MSS give 90 to 98 stanzas. 43. The Br. Mus. MS appears to give 99 stanzas, but the number is really 100; for the consecutive twin verses analpa-cinta-bhara-and iti priye prcchati are both marked by the same number 76. The actual number of stanzas given by the chief commentaries is as follows: Arjuna--102; Caturbhuja--107; Koka--100; Surya--101; Vema--101; Ramananda--101; Ravicandra 100; Rudrama--100 in Br. Mus. MS (114-115 in BORI MSS); Ramarudra--106. 44. But he suspects two of his included stanzas as interpolated (Nos. 59 and 71). 45. Kosambi, preface, p.3. 46. ZDMG, xxvii, 1873, p.7f. 47. The metres next in order of frequency are Harini--16, and Sikharini--12. The other metres occurring in the threee recensions are: Vasantatilaka (as many as 11 instances in the E recension), Sragdhara and Mandakranta (3 or 4 st.) and Malini (2 st.). Only two instances of Drutavilambita and Vamsastha occur in E and W recensions respectively.--Aufrecht's suggestion is highly plausible, but unfortunately the MS tradition does not support it. 48. See Winternitz, Geschichte, iii.p. f.n. 49. Most of the earlier fine erotic verses have a tendency to find thier way into the Amaru-sataka or to be ascribed to the poet. 50. Caturbhuja, however, appears to be aware of Arjuna's rejection of these stanzas as spurious for he says : atra kecin mudrita-pamsava ity arabhya pito yata iti paryantam praksepah sloka iti vadanti (fol. 26b). 51. With regard to Ravicandra's commentary itself, we should note here that it has not yet been properly edited. We have taken the text of the Calcutta ed. of 1808 (100 stanzas)as the basis of our study; but on consulting MSS we find that the text of Ravicandra, like that of Rudrama, exists in diverse form and extent. The India Office MS No. 4003-4/1392a and b (Eggeling, Catalogue, vii. p. 1520f), in Bengali characters, first gives the text only in 98 stanzas, terminating abruptly without colophon. But the commentary which follows separately comprises 95 stanzas in the order and number in which they are given in India Office MS 4005/711b as described below, and not in the order and number in which they are found in the accompanying MS of the text! The commentary contains a colophon. Another Inida Office MS (No. 4005/711b), also in Bengali characters, is a curious appropriation of Ravicandra's commentary by one whose name or rather title is substituted in one of the opening verses as `Vidyavinoda-sukrti' for Jnanananda Kaladhara, although the substitution (in somewhat paler ink) makes the Padametrically defective! The change, however, is deliberately confirmed by the colophon, which describes the author of the commentaryas purvagrami-kula-kalanidhi-sri-vidyavinodacarya-bhattacarya. This MS (==Simon's Cw) contains comments on 95 stanzas only. Up to its st. 79 (drstah katara-netraya) it follows the order of the printed text, but thereafter the stanzas are arranged differently, and three new stanzas are introduced (for these see Simon p. 134, Nos. 93-95). One other India Office MSS(No. 4006/711a) which we have already mentioned above, also in Bengali characters (==Simon's Cy), carries the text up to st. ayate dayite, counted as No.90 in the MS. The BORI MS (No.458 of 1891-95; Gode's Catalogue, xiii, pt.1,14) is in Devanagari, but it is manifestly incomplete, breaking off at st. 68 (pito yatah prabhrti). This appears to be the same MS as noticed by Kathavate in his Report 1891-95 (Bombay 1901),p. 14. Stein's Jammu MS appears also to be fragmentary and contains only 25 st. (Stein, Catalogue of Skt. MSS at Jammu, Bombay 1894, No.560, p. 66; p. 277). The complete MSS, however, not only in extent but also in readings, differ very considerably; and a critical study of these and other available MDD is necessary because of the divergences of the respective texts presented by them. The same remarks apply to a certain extent to the unedited text of Rudrama; while the single alleged MS of Ramarudra can hardly be accepted without reservation.--A critical edition of Rudrama's unpublished commentary with the text, based on 5 MSS, is completed by the present writer, and will be published in the next issue. 52. This well known stanza (see our Index) is also assigned to Vikatanitamba in some of the Anthologies. 53. Assigned to Valmiki-muni in Sp. 54. Anonumous in all citations. This st. however, is not commented upon by the MS which includes it; possibly it was originally an illustrative marginalia (giving a parallel st.) which was subseqyently absorbed into the text. 55. A comparative table is given below of the stanzas found in the three versions (Ravicandra, Rudramadeva and Ramarudra) respectively to show the discrepancies in the choice and order of stanzas in the Eastern recension. 56. Even the Dasa-rupaka, one of our earliest testimonia (end of the 10th century), which otherwise correctly assigns stanzas to the Amaru-sataka by name, cites the two verses puras tanvya the Southern and in the other by the Southern and Eastern recensions respectively !