Goddess Vinayakf ] [ 9

In certain Puranic texts the goddess is found mentioned not by thename VinayakI or its variants but only by qualifying epithets based onher elephant-headed form. The occurrence of such goddess names indifferent texts may not be always taken to indicate the existence of afully evolved VinayakI aspect as the counterpart of Ganesa. In thesecases texts appear to offer evidence for the rudimentary belief in anelephant-headed Matrka, who had emerged in the religious myths likemany other animal-headed and bird-headed folk goddesses. Such afemale being might have served the prototype to the later VinayakI figure.The Bodhayana Grhya-parisistas, the terracotta plaque from Rairh andthe Mathura Matrka-relief bear indirect testimony presumably to thesame fact.

In the Harivamsa the Mothers ( mdlarah ) ^re not only mentioned bytheir individual names but a general description of their features andornamentations is also added. This includes a reference to their com-posite aspect as having animal and bird heads, such as that of cats,rhinoceroses, elephants, lions, etc.42 Thus a goddess or goddess-typewas believed to have an elephant-head. Similarly, the elephant-headedgoddesses ( in plural ) are described in the Vayu Parana along with manyother such female beings staying in the abode of Siva.43 In the Nagara-khanda of the Skanda Purana one Gajamukhi goddess is mentioned asthe first among those innumerable Matarah who were born at Hatake-Svara from a sacred Kunda to help the two widowed wives of the kingof Kai against Kalayavanas.44 Further in the same Purana is found themyth of Laksmi assuming an elephant-head (gajamukhi, gajavaktrd) dueto the curse by a displeased Brahma na.45

In the Kd&i-khanda of the same Purana is described a Gajavaktrdamong the multiple goddesses created by the Great Goddess to assist herin her fight against the demon Durga.46 Again, the goddess Gajdsya is saidto be one of the Nine Saktis, who are presiding divinities of the directionsand protect the Kasi region on all sides.47 However, the name Gajamukhd

42. II. 109. 54.43. 101. 262.44. 88. 23.45. Ch. 81, particularly verse 17; Ch, 85.46. 72. 5.

47. 72. 90. The other names include Satanetra, Sahasrasya, Ayutabhuja,Tvarita, havavahim, Viiva, Saubhagyagauri.

1 ] [The Female Ganesa

is also counted in the list of 1008 Names of the goddess Lalita ( Lalitd-sahasrandma) given in the JVdrada Purdna.*8

Of outstanding significance is the evidence we have in the JVdradaand the Brahmdnda Purdnas that give lists of more than fifty aspects orforms of Vighnesa-Ganesa each of whom is described as having his ownSakti, or female counterpart,49 Similar lists of fifty Ganapatis and theirfemale eaergies are found collected in the texts such as the JndndrnavaTantra, Tanlra Kaumudi, Ucchistaganapatyupdsand, and the PaddrthddarsaCommentary on the Sdraddtilaka Tantra.

In these lists are found such goddess names for the Ganapati-Saktisas Hri, Sri, Pusti, Sarasvatr, Kali, etc. besides VighnesJ, or Vighnesanl.One can hardly be sure as to the bodily aspect of these female divinitiesand it is difficult to assimilate any one of the names with the goddessVinayakI except perhaps VighneSi or Vighnesanl, which is the only namein the long lists that appear to have been derived directly from a regularepithet or name of Ganesa. In both of the Puranas it is stated thatGanesas should be shown as having an elephant-head with three eyes,holding in their four hands a pdsa ( noose ), an ankusa ( goad ), varada andabhayamudrds and be accompanied by their respective Saktis.51 In theJVdrada Parana the Saktis of Ganesa are significantly called Vighnesamdtrkd,or Ganesa-mdtrkd, and the presiding deity of this group is Sakti-ganesvara;62

the consort of Ganapati is specifically said to carry a lotus53 although anyfurther information regarding her iconography is wanting.

It may not be out of place here to draw attention to the icono-graphic form described by Srikumara ( 16th cent. A.D. ) in his Silparatnaunder the name "Sakti-Ganapati" : At ha Sakti-Ganapatih :dvdbhydmvibhrdjamdndm drutakanakamahdsTnkhaldbhydni kardbhydm Bijdpurddiiumbha-dda&abhujalalitam pancabljasvarupam / sandhydsinduravarnam stanabharanamitam

48. I. 89. 119.49. See Appendix I.5t). Also see Pranatosinl, 3rd edn, Calcutta 1898, pp. 83-4.

51. (a) Brahmanda P. II . IV.44.64: .* taruti'drunasainkatan gajavaktrarlstrilocan&nfia:iankiiavrfrabhitihastancliaktisatnanvitan.

(b) Ibid verse 77 : varadabliayxha&tadhyanchaktyaliiintcivigralian.

(c) Narada P. 1.06. 1H9 ; fiasankutabhayavaran dadhattain kartjahastayajpatny~d-histctm raktatanum trinetram gctnape. bhctvet.

52. I. 00. 137-38.53. katljahastayV pa

Gcddess VinayakI ] [ 11

iundilam sanniiambam Kanthddurdhvam karlndra yuvatimayamadho ta*nnaumi deva (ganesamjj

"Now Sakti-Ganapati. Salutation to the god Ganesa, who is fromthe neck upwards an elephant and whose lower body is of a youthfulfemale, has vermilion-red colour of the evening, a corpulent belly, thebreasts which make the bodily frame bow down owing to their weight,and beautiful hips; is shining with two proboscises which are ( like ) twogreat chains of molten gold, is of the form of Five Seed-syllables (panca-bijasvarupam ) and has ten splendid arms adorned with a citron ( bljapura ),etc. attributes."54

In fact the above form of Sakti-Ganapati is given in the text underthe category of syncretistic icons ( misra-murtayah ) along with the imagesof Hariharavidhi, Laksmlnarayana, Ardhanarisvara and Samkarana-rayana.55 As the text cited above is in itself apparent and there is,however, no confusion in its literal meaning, the interesting image of acomposite form of Ganapati and his consort described therein is beyondquestion. According to the text the blended icon of the god and hisSakti is characterised by the following features :

(1) in its portion above the neck it is elephantine;

(2) the lower body is that of a youthful female, having breasts andbeautiful hips;

(3) it has two proboscises, apparently one belonging to the god andthe other to the Sakti;

(4) it has ten arms; the attributes held in the hands are not givenexcept for a citron.

This syncretic concept appears unique in its type and we are notaware of any other literary account describing it in such details; nor isany image known to us which would tally with this iconography.However, we may note for the present that the author of this iconogra-phic text very well knew a female form of the god Ganapati that wascharacterised by an elephantine head similar to the male god. This fact

54. IIXXV. 74 (Trivandrum Skt Series edn, 1929, p. 162).However, the English rendering of this verse as given by D.C Bhattacharyya ("TheComposite image of Vasudeva and Laksmi,"JAS, Calcutta, IV. Series, VII, No. 4,1966, p. 265 ) is hopelessly incorrect. See notes 7779.

55. Silparatnall. 25. 73, 7579.

12 ] [ The Female Ganesa

is apparent from the mention of two elephantine proboscises in theircomposite figure.

Perhaps the only early text56 affording some side-light on the iconof the goddess Vinayaka is the Buddhist Aryamanjusrimulakalpa*1 It isstated that the region of the Vindhya is appropriate for the siddhi ofVinayaka. She is the deity causing obstacles to those who mutter reli-gious charms. She possesses the elephant-proboscis (hastdkdra-samdyuktdm )and is one toothed (ekadantdm58). She is of the form of a horse ( aivdrupd )and can assume many different forms. She is the divine daughter oflana, is of varying nature and causes obstacles ( vighnakdrakdm ).

Thus we are informed that the goddess Vinayaka was an ill-natureddivinity worshipped in the Vindhyan region in her own individual capa-city. She was presumably a female transformation of the god Vinayakaand hence possessed an elephant-head, one tooth and was believed to bethe daughter of Isana-Siva just as Ganesa was his son.

The goddess appears to have had only a secondary status in thepantheon although sometimes she did enjoy quite an independent perso-nality from Vinayaka. As such she is found reckoned as one of theSixty-four Yogini goddesses both in the Yogini Lists preserved in literatureand in the artistic representations of Yoginls. It is under the nameVinayaki or Vainayaki that we find her in several Causath-Yogini-Namavalis published by A. C. Nahta69 and V. S. Agrawala.60 Her placein the above lists is either 25th or 33rd. Similarly, under the nameVainayaki, she is the 36th in the list of Sixty-four Yoginls enumeratedin the Bhagavatl-mandala incorporated in the Acdradinakara ( completed

56. B. Bhattacharyya ascribes this text in its original form to the second century A.D.

( see Guhyasamnja Tantray GOS, vol. LIU, Baroda 1931, pp. xxxiii ff. ). However,the text is not later than the 7th-8th cent. A. D.

57. XXX. ( Trivandrum Skt Series No. LXXVI, vol.11, 1922 ), p. 326 : siddhir vindydkd

tatra vighnakartd sajctpiriftm / hastak&ra&amayuktflnekadant&tn mahaujasUm //atvarup'a tathanekakaratalitiatn / Isanasya sutam divyatn vividham vighncr-karakam //

58. It is the amended reading suggestible in place of the printed one. eynkfanekae

should be QyuktUmzkaQ .

59. "Bavanaltra Namavalt^ in todha-patrika, Year 14, No. B, 1963 July, Udaypur,

pp. 191-94.

60. Praclna Bharatiya Lokadhamict, pp. 147-49; Ancient Indian Folk Cults> Varanasi

1970, pp. 204-06.

Goddess Vinayaki ] [ 1 3

in VS. 1468 ) of Vardhamanasuri.61 In the Sixty-four Yogini namesrecorded by the great Jaina Mantrika teacher Jinaphrabhasuri in hisfamous Vidhimngaprapa ( composed in VS. 1363A. D. 1306 ), her nameVinayaki is at the 14th place.62 But in several other lists of the Yoginisshe appears under the name Ganesvari and in the light of the abovecited lists there can be no doubt that the same goddess is intended bythis name Ganesvari. In the Catuhsasfi Yogini List based on a manus-cript in the Hamsavijaya Collection of the Jaina Mandir, Baroda, givenby Bhattacharya in his Jaina Iconography, her position is the fourth inorder;63 the Catuhsastiyoginistotra of an unknown author, included asAppendix XIV to the Ahmedabad edition of the Bhairava-PadmavailKalpa^ gives the name Ganesvari at the fifth place, while she occupiesthe place of honour being the first name in the list of Yoginis, lauded inthe Goddess Hymn ( Devi-stotra), given as Appendix XV of the aboveedition of the Bhairava-Padmdvati Kalpa.*5

The Sixty-four Yogini names given in the Ka!-khanda have hermentioned by the epithet Gajdnana, Elephant-faced, and she is indeed reck-oned first in the list.66 The presence of an elephant-headed goddess in theextant images of the three Sixty-four Yogini shrines in Madhya Pradeshand Orissa goes to reveal that she was regarded a regular member of theYoginis during medieval times. The Vinayaki sculpture from Hingalaj-garh ( literally, the Fort of Goddess Hingulaja ), district Mandasor, alsoappears to belong to a Causath Yogini temple now obliterated.67

From the several Jaina lists of Yoginis, cited above, it is conceivablethat the name Ganesvari was also of common occurrence for denoting thesame Yogini goddess Vinayaki.68 Thus, we find, in the Vanadurgopanisad thegoddess, now under the name Ganesvari, invoked along with the MatrkSs,

61. Ch. 33 ( vol. II, p. 207 ).62. Ed. Jinavijaya, Surat 1941, p. 116.63. 2nd edn, Delhi 1974, p. 137; also see p. 16.

64. M. B. Jha^ery and K.V. Abhyankar, Comparative ami Critical Study of Mantra-iastra, Being the Introduction to Sri Bhaiiava Padmavati Kalpa, Ahmedabad1944, p. 101.

65. Ibid., p. 83.6T Skmda Purana IV. 45. 34,67. See below, and Ills. 10-13.

68. But see Padma Parana, Svarga Khanda, 35. 9, wherein the word a$elvarl isused to signify a female gaiia of iva.

14 ] [ The Female Ganesa

Dikpalas, their female aspects, Yoginfs, etc. and associated with Siva.69

She is worshipped to protect the devotee on all sides by her favouringhand that bears a parasuS0 The parasu or battle-axe, however, appearsto have been understood as her characteristic attribute, similar tothat of Gaaapati, and it is usually found held in one of her hands inmost of the sculptures of this goddess known so far.

The association of this goddess with Siva through Ganapati canhardly be much emphasised. The inclusion of name Lambodari, appa-rently derived from another regular epithet, Lambodara, of Ganesa, in thelist of Fifty Saictis of Rudra in the Hdnasivagurudevapaddhati is thus signi-ficant.71 Similarly, the goddess Lambodari as the female associate ofdiva's Ekanetra form in the Srikanthadi-nyasa enumerated in theBrahmanda Purdna, may be taken to specify the same goddess Vinayaki.72

We have not been able so far to trace any passage in the vast Indianliterature which may specifically include goddess Vinayaki among theSeven or Eight Mothers ( Astamdtrkd ). Nevertheless, a prose passage givenin the Skanda Purdna concerning the sddhand of Mahavidyas consists of aninvocation to goddesses of different nature in which Vinayaki is placedamidst the well-known names of the Astamdtrkas. However,the Gadhwa panel ( ///. 7 ) showing goddess Vinayaki with figures ofVarunani and Narasimhi, and the fact that the particular slab appears tobelong to a set of three panels depicting Eight Matrkas with Ganesa, maytend to substantiate a view that in some areas of her worship she wasgiven a place in the group of Eight Mothers.

It is a happy chance that the above observation can, however, befurther supported by one of the old Sanskrit documents preserved in theisland of Bali. In the Balinese ritual text called Vedaparikrama, edited bySylvain Levi in the Gaekwad's Oriental Series ( No. LVII ), the invocationmade to the goddesses includes seven names of Brahmi, Mahesvan,KuTturi, MaheSvari, IndranT, Ganendri and Vaisnavl :74 0 Brahmi vicetsiohd. purvdm o**1 mahesvan vicet svdhd. daksind. om kumdrl vicet sidhd. uttard.

09. C. Kunhan Raja, ed., Unpublished Upanisads, Adyar 1933, p. 453.70. Om namo bhagavati Ganeivari parasu-hastena sarvato mam raksa raksa \

71. I. II. 11.72. II. IV. 44. 59.

73. M-ihesvara Khand* ( I ) 02.00. Brahmani Waheivari Varalii Vinayaki AindriAgneyi Camunde Varuni . . . .

74. Balidvlpagranthah, Baroda, 1933, p. 12.

Goddess Vinayaki ] [15

0 Mahesvari vicet svaha. agneya. o Indrdnl vicet svaha* w&anya. o Ganendrl

vicet svdhd. vayavya. o Vaisnavi vicet svdhd. pascima. . . .

The name Ganendrl here appears to refer obviously to the femaleaspect or form of Gamndra or Gauapati. It is further remarkable that sheis associted with the direction presided over by Vayu.

We may mention here another significant myth concerning the roleof an elephant-headed Raksasi in the story concerning the birth ofGanapati. In the Haracanta of Jayaratha, who lived in Kashmir in the13th century A. D., the account given of Ganesa's birth describes theelephant-headed demoness named Malini, who dwelt at the mouth of thesacred river Ganga, as drinking the bodily impurities mixed with unguentrubbings of Parvati to give birth to a male child with five elephant-heads,whom Parvati and Siva subsequently claimed as their offspring.75 Thiscurious legend is not found in any of the Puranic accounts in which noagency of such an elep'iant-headed female was thought to be essential.

As noted above, the Silparatna of the Keralese Brahmana Srikumara( 16th century A. D.) describes a composite form of half-female half-male Ganesa image under the name Sakti-Ganapati. This particularsyncretic icon of Ganesa was apparently evolved on the lines of the well-known Ardhanarlsvara form of Siva combining the bodily form of hisfemale Sakti Ambika. The motif is explicitly stated in a poetic fancyfound in the first verse of the Halayudhastotra inscribed in 1063 A. D. inthe AmreSvara Temple at Mandhata, Mimar district, Madhya Pradesh.77

It is further believed that the dual form *ganipatayoh? in the BaijanathaPrasasti78 of Saka 1126 ( =1204 A. D. )79 may also be taken to allude tothis composite image of Ganesa. However, one thing is undoubted thatthe blended form of Ganapati with his female,, as given by the Silparatna,affords the testimony to an already existing belief in the female aspect ofGanesa as Vinayaki, the female goddess "Ganapati". As the goddessVinayaki and the male god Ganesa were known to the worshippers tohave identical iconographic forms of elephant-faced divinities apart from

75. Ch. XVIII. 5 ff.76. See above pp. 10-11.

77. Manisha Mukhopadhyaya in J. of Ancient Indian History, Calcutta Uni., vol. I I ,

pp. 112-5.

78. Ep. Ind., vol. I, pp. 112 ff.

79. This is tlie date corrected by D. C, Sircar, see his "Indological Notes", JAIH*.vol. IV, pp. 181-6.

16 ] [ The Female Ganea

the sex, the fused image of theirs in the Sakti-Ganapati form is describedto have two proboscises, one of the goddess and the other of the godhimself. Further, it may be remarked that it is the icon of the male godcombining the bodily form of the goddess in his left half side, just similar inpattern to the ArdhanarTsvara Siva. But in case of Ardhanarl GaneSa, theelephant-faced head of the female cannot be simply differentiated fromthe male features; thus, the form below the neck is said explicitly to beportraying the female attributes, as consisting of the form that is of ayoung girl (yuvati-mayam ). Unfortunately, no such blended image ofSakti-Ganapati is known from sculpture and we can not be sure as to thedevice followed by artists in delineating the female bodily portions insuch an image. Nor do we have any other specimen of a similar casewhere the blending of two animal-headed divinities is to be achieved inone single image. In the Silparatna, the several expressions describingthe lower portion of the image are indeed liable to be interpreted in twoways as follows :

( 1 ) Kanthdd adho yuvatimayam. Whose body below the neck is that of ayouthful female; or consists of partially ( the body ) of a youthfulfemale.

( 2 ) Stanabharanamitam. Who is bent owing to the weight of the breasts;or of a ( single ) breasU

( 3 ) Sannitambam. Who is endowed with ample buttocks; or an amplebuttock.Thus it is possible that the poet was not giving a form in which

the complete lower body was that of a female, but he had before him theimage in which only the lefc half was endowed with female attributes.The image is descrided to have ten arms in one of which a byapura ismentioned. We do not know what other nine attributes were, althoughthe list often attributes held in the ten-armed image of Ganapati can beknown from both the texts and sculptures.80 There is no indication tospecify whether in the ten arms, five held attributes characteristic tothe male god and the other five to the goddess, although such a divisionconceptually is not unlikely.

80. See for example, Silparatna, part II, pp. 166, 159; lsana$ivaurudevapaddhati,1.16.7-8; Brahtriandu Purana II. IV. 27. 69-70 (but the attributes given aremore than ten); Skanda Purana I. I. 11.9-10; Getty, Ganeia, pi. 18, c; 19;Haridas Mitra, Ganapati^ pi. facing p. 60; etc.

Goddess Vinayaki ] [ 17

It may not be out of place here to observe one further significantpoint concerning the nature of the goddess Vinayaki. One has to bearin mind that the iconographic and mythological problems concerningVinayaki are not isolated from those of similar animal-faced goddesses.Such female beings must not necessarily be confused or confounded withother general goddesses or human-faced consorts of deities. For thepresent instance, Gaiiapati himself is well known to have his normalconsort-goddesses in Buddhi, Siddhi, and sometimes, Astasiddhis,Rddhi, etc.,81 and many other names enumerated along with his multipleaspects.82 But a Vinayaki or a Varahi is not simply the spouse orSakti of Vinayka or Varaha. Although the origin or model of such agoddess might have been inspired or derived from the male god,the goddess Vinayaki or Varahi once evolved on the type of the maledeity concerned, existed in the pantheon quite as an independent entity.According to the well stated opinion of the Puranic writers, the seven oreight great Matrkas originated with the great gods themselves assumingfemale aspects to aid the Great Goddess in her exploits against demons.83

They are indeed described to be identical in form to their male counter-parts, and bearing specific attributes and outstanding cognisances of theiroriginal male aspects.84 In fact, in these cases the male god was believedto have transformed into a goddess, thus the names being onTy simplegrammatical forms in feminine, viz. Varahi from Varaha, N irasimhlfrom Narasimha, KaumarT from Kumara, IndranT from lndra, Brahmifrom Brahma. For the specific iconographic aspect, the goddesses ofanimal-headed type appear to have never evolved much beyond theoriginal deity. This presumbly is the reason that usually no such imagesare found where a god of this list is shown as accompanied by his femaletransformation.85 As far as our knowledge goes at present, we do nothave any such early instancs of an image in which Ganesa is depicted as

81. Getty, op. cit., pp. 35-(V, G. S. Gliurye, GoJ.s and Men, Bombay 11T.2, pp. 78-0.

82. See Appendix I.

83. Sivamaliapurana II. V. 4:>. 24-25.

84. Bralunanda Parana II. IV. 19. 9-10.

85. In the Bheraghat Yogini figure of Vinayaki the elephant-headed god underneath herfeet is of a subordinate status. For similar form of Varahi riding Varaha, seeBhavisya Pur ana I. 177. 11.

18 3 [ The Female

having elephant-faced VinayakJ as his wife, or Varaha with boar-headedVarahi, or Narasitpha with Narasimhi. In the sculptures whereverGanesa is seen with his wife or energy she is not an elephant-headed beingbut of a simple female form.88 However* we are not to take into consi*deration here the very late Tantric Buddhist representations of Sakti-Ganapati or his Yab-yum icons sometimes showing both the male andfemale in identical forms*87 It may be observed further that the formssuch as Varahi, Narasimhi* Hayagrivl, Vinayaki* etc* had basically evolvedunder the Tantric formulations and were considered independent godde*sses, while the identity of the regular consort-goddesses of the majorBrahmanical gods was a different matter ( e,g. Varaha has Prthivi theEarth-goddess, Brahma has Sarasvati% Kumara has Kaumari, Visiju hasLaksml or Vaisnavi, Vinayaka has Siddhi or Rddhi, all these havinghuman-faced normal goddess-form ). Particularly in the case of suchgods only the above phenomenon is readily recognisable beyond confusionwhere hybrid or composite human-and-animal form is involved in th$iccnography*

86. The earliest of such representations of Gnnes^ with his human-faced consort is of theGiipta period from Bhumara. In medieval iconography there are many siicfosculptures; see Getty, dp. cit., pi. 3 a; 4 a and &, 14 e, 10.

$7. Ibid., pp, 72 ff,

IV

ELEPHANT.FACED GODDESS VINAYAKFIN INDIAN ART

In the foregoing discussion we have made reference to several scul-ptures portraying the goddess Vinayakl. A detailed iconographic analysisof all the representations so far known to us of this goddess is likely toshed further light on this obscure divinity. Although it is a hard factthat we have not been able as yet to find a text which describes the iconof the female Ganapati in its full relevant details of attributes in handsand other characterising traits, the identity of the goddess in sculpturalrepresentations is readily revealed from her essential elephant-headedaspect. However, from the texts cited above the following characteristictrails of her icon can be gathered :

( i ) Her elephant-head similar to the model of Ganapati. This isto be an invariable peculiarity of her image, like the original Ganesa,mentioned in the texts by epithets such as Hastimukha, Gajamukhd, GajdnanaGojavaktrti, or names derived from those of Ganesa such as Vinayaki,Ganesvari, Lambodarl, Ganesl, Vighnesi, "Ganesdni".

( i i ) Possibly, like her counterpart Ganapati, she was to have onlyone tusk; see Aryamanjusrimulakalpa.88

(iii ) The same text in its present corrupt form reads the word"asvdrupdm" as one of her adjectival phrases. Can we emend it as"aivdrudhdm" and deduce that in one of her representations she wasshown as riding a horse ?

( iv ) In one text she is said to carry a parasu ( battle-axe or hatchet )in one of her hands.89 In the earliest of the images of god Ganapatiknown so far he is invariably shown to bear a hatchet in one of his twohands.90 Thus it appears to be a significant trait which was presumablyadopted in the iconography of the goddess Vinayaki quite early fromher model.

88. Above p, 12.89. Above pp. lo-4.90. J. N. Banerjea, DHL, p. 357.

20 ] [ The Female Ganes'a

The archaeological evidence in hand presently shows the figure ofan elephant-faced female already made by an artist of Rajasthan aboutthe first century A. D. or slightly earlier. It sounds rather strange inface of the fact that no extant images of the elephant-headed god himselfcan be ascribed to such an early period. According to the verdict ofseveral authorities on the subject, the full-fledged image of Ganesa hadpossibly not evolved before the fourth century A. D.91 But this opinionneeds now to be revised in the light of a fragmentary relief from Mathuraof the late Kuslna period ( ///. 1 ) , and particularly on the basis of theelephant-faced female terracotta figure from Rairh. Moreover, now thereis ample literary evidence to show beyond doubt that the myth of theelephant-headed god had already come to be popular about the firstcentury A. D., if not earlier.

I. Terracotta plaque, foundduring the excavations at Rairh, informer Jaipur State.92 Stylisticallydatable from 1st cent. B.C. to 1stcent. A. D., it shows a female figure( Textfig. I; 111. 5 ) with an elephant-facestanding in frontal posture. Owing tothe rolled state of the piece the detailsare much obliterated, but she appears tobe wearing a sort of high headdress orcrown. There seems to be htld someelongated object in the right hand ofthe figure. Are we to recognise thisattribute to be a hatchet ? If so, theidentity of the figure carrying thischaracteristic attribute of parasu can beascertained aa that of Vinayaki. However,in the light of the Jyestha KalpaoftheBaudhayana Grhyaparisistas, she may bedescribed by the name HastimukhaJyestha. It may be observed that thereare several Kusa^a style reliefs from Elephant-faced Jyestha.

From Rairh

91. ft. G. Bhandnrkar, Vtrisnavism, Saivism and Minor Reli*:ous Systems CollectedWorks edn, Poona 1929, pp. 211-2; Getty, op. cit., p. 10;* J. N. baneriea, op citp . of>9. J i J

92. K. N. I tiri, Excavations at Rairh, p. 29, pi. XIV, fig. No. / .

Goddess Vinayaki ] [21

Mathura, preserved in the Mathura Museum ( Ills. 3-4 ) , which showa group of seated female figures or Matrkis endowed with faces ofbirds and animals including goat, owl, parrot, bull, boar, lion, etc.91

Earlier scholars have so far failed to recognise any instances of anelephant-faced Matrka in such reliefs of animal-faced goddesses fromMathura. It is, however, presumable that a goddess with elephant-facemight^have evolved under the same mythical formulations. In onespecimen of the Kusana period Matrka reliefs, we have venturedbelow to identify a goddess figure having the face of an elephant,

II. Fragmentary relief from Mathura, showing a row of seatedMatrka figures ( ///. 4 ). Ht 5 inches. Mathura Museum, No. 2331.The surviving panel shows five of the original seven Mothers, enchseated in bhadrdsana and bearing a child in the lap. Their right hands areheld in abhayamudnl. Their iconographic distinction is based primarilyon their animal and bird heads.

For our study here, the figure on extreme right is of special signi-ficance ( Frontispiece ) Her head can be recognised to represent th&t of anelephant from the trunk turned to her right as curved inwards and ihelarge elephantine ears. To adjust the proportionately large elephant-face the artist has shown it in profile and the proboscis as incurvedin the lower portion. This archaeological evidence, together with theRairh terracotta, goes to testify to the early emergence of an elephant-headed goddess who had come to occupy a pi ice in the group ofDivine Mothers already in the Kusana period iconography.

III. A very interesting group of the Matrka panels is preserved bythe Archaeological Survey Department in their site collection at Gadhwa,near Allahabad. One of the slabs ( ///. 6 ) shows a seated figure ofthe elephant-headed goddess Vinayaki along with those of goddessesVarunanl and N irasimhi. This rectangular stone panel (measuring about42 inches by 26 inches ; 1-065 m. by O660 m.; No. RK 7 ), thoughpreserved with several deplorable injuries, appears to have formed partof a set of three such slabs depicting in all nine figures, eight of theMatrkas and one of Ganesa along with them. The other stylistically

93. V .S. Afrawala, in JUPHS., Ol:l Series, vol. XXII, 1919, pp. 101-3, Museum \

22 ] [ The Female Ganesa

similar panel belonging seemingly to this set is numbered RK 6 in thecollection and shows Camunda and Varahi with Ganesa poised inbetween them. The third slab not to be seen in the site collectionnow might have represented three more goddesses of the group, fromprobable names of Kaumarl, Brahmanf, Mahesvarl, Vaisnavl or Indrani.

If the above observation made by us is correct the significance ofthis portrayal ofVinayakl included in the group of Eight Matrkas isgreatly increased. Probably some side light is shed on this problem bya passage in the Skanda Purdna.91 She was no doubt included in the Sixty-four YoginI lists, as noted above, but here we appear to have herrepresentation as a member of the Eight Matrka group, indeed a veryrare occurrence of this goddess paralleled, at present, by no othermedieval sculpture.

The four-armed goddess is seen seated gracefully in the lalita posture( ///. 7 ). Her bodily form is revealed as almost nude with no trace ofdrapery above a beaded girdle but she appears to wear below a close-fitting diaphanous garment of which the fringe is seen in incised detailsover the legs. Other beaded ornaments are worn on the wrists, roundthe neck and on the elephantine head. The bulky yet charmingly rotundfleshy limbs of the body make an extremely elegant figure. Stylisticallythe sculpture may safely be dated around 800 A. D. The elephant-headhas only one tooth characterising the ekadanta trait of the Ganesa icono-graphy which appears to be mentioned also by the Aryamanjusnmalakalpain the description this goddess.95 The proboscis is turned properly to theleft side of her and reaching the bowl of sweets, held in her lower lefthand. She is four-armed and in the upper left hand she carries a mace( kuta ) shown vertically. Further, she holds a battle-axe ( parasu ) in thefront right hand and a serpent in the back one. The presence of parasuas her characteristic attribute is substantiated by the Vanadargopanisad**and this as well as the serpent are similarly familiar attributes of theGanapati images. Below the seat is represented the figure of her vehicle,Mouse. Thus we have in the present sculpture a fully evolved early medi-eval type of Ganesa's ic3n transformed into a goddess figure with youthfulfemale body.

94. See above pp. 14-15.

95. See abov'e p. 12.

9G. See above pp. 13-1-i.

Goddess Vinayaki ] [ 23

IV. Stone slab from Suhania, Morena districts Madhya Pradesh,Now in the Gwalior Museum. C. lOih century A.D.97 ( ///. 8 ).

The goddess is shown inside an architectural frame of two pilasterswith two figures of griffins ( vydlas ) decorating their outer side. A femaleattendant carrying a lotus-stalk is seen standing below each of the mythi-cal creatures. The main figure of the goddess inside the niche is standingin tribhanga posture-, clad in a knee-reaching lower garment but wearingon the bare upper body only ornaments including a pearl-necklace dang-ling between the rotund breasts, armlets, bracelets, a garland and akaranda-mukuta on the elephantine head. Unlike the Gadhwa figure sheis pot-bellied in a manner as is usual with the icons of Ganesa. She isfour-armed conspicuously carrying in her normal right hand a parasu andin the additional upper hand a lotus-bud^ In the upper left hand she isholding possibly a dagger which is partially damaged ( this attribute maybe even a serpent or some other object ), and in the corresponding normalhand is borne a bowl of sweets^ which is reached by her proboscis turnedsharply to the proper left. The presence of two small figures of maleattendants near ier feet, a drum-beater on her left and a flute-player onthe right, appears to signify the mood of the representation of music anddance. Presumably, the flexed bodily attitude of the goddess suggests thesame gay dancing situation. The specimen does not appear to have be-longed to the series of Sixty-four YoginI goddesses for there is no suchtradition recorded in case of its place of origin, and the treatment of thesculpture is quite \inlike that of the Vinayaki figures known from CausathYoginI complexes of Madhya Pradesh and Orissa.

V. Fragmentary sculpture of Vinayaki, embedded in wall of a Sivaftetnple, known locally as Purana Mahadeva, at Harshagiri, near Sikar,Rajasthan98 (Textfig.2). About 10th cent. A. D. Only her elephant-head and the torso up to the hips is partially preserved. It is difficukto ascertain whether the figure was seated or standing, or had only two ormore arms. However, she is bearing in the hand of the left arm bentto the front a bowl, while the right arm is damaged from the elbow.The head, backed by a lotus-halo, wears a high karanda-mukuia of tieredgarland with a jewel on the front. The figure being thickly coated withvermilion-paint is rather obscured in its finely carved details. However,

=97. B. N.Shanna, in Oriental Art, vol. XVI, No. -2, Summer 1970, p. My, fig. 3.98. Mary C. Lanius, op. cit., p . i>8.

24 ] [ The Female Ganesa

she is adorned with a multi-stringed pearl-necklace, a thick garlandcoming over the ornate armlets. The remarkable feature is the usualturn of the trunk to left which reaches near the bowl held in the lefthand. The figure is of corpulent type with full-rounded heavy breastsand a bulging belly, and stylistically appears to be an impressive piece ofRajasthanese medieval sculpture. There is nothing possibly to suggestthat the figure could have beioned to a Causath Yogini group.

r

L Vinayakl. From Harshagiri.

10th century A. D.

VI. Image stele from Giriyek, Patna district, Bihar, now in theIndian Museum, Calcutta, Reg. No. 3919," R. D. Banerji commentedthus on the sculpture : "An unique image, the Sakti of Ganesa, is alsoto be seen in this collection. It is an elephant-headed goddess, seatedwith a rat on the pedestal "

It is a Pala sculpture belonging stylistically to the 1 Ith century A. D,( ///. 9 ). The icon details are, however, well presen ed showing the tour-

99. T. Bloch, Supplementary Catalogue of the Archaeological Collection of the IndianMuseum, p. 90; P. D. Banerji, Eastern Indian School of Medieval Sculpture,

p. 117; B. N. Sharma, "A rare Image of GayCsani ^ r o m Giryak ( Bihar)," JBRS.,vol. LVI, 1970, pp. 50-52.

Goddess Vinayaki ] [ 25

armed elephant-headed goddess Vinayaki, seated in the proper padmdsanaposture on a seat of spread-out lotus, placed on a highly ornate pedestal.The figure represents an elegant female form with no feeling of thecorpulent bodily bulk remarkable in other images of this goddess. Sheis clad in a close-fitting lower garment tied with a sash round the girdleand reaching just above the ankles. The ornaments worn include armlets,wristlets, a torque and an elaborate s:ring on the high elephantinehead. A remarkable feature is the strip-like long sacred-thread.She is holding a battle-axe in the upper left hand and a conch-sheil inthe normal left hand, placed on the thigh. In the normal right hand israised a staff and in the extra right hand a small vase. The Tantriccharacter of the representation is evide.it from her sitting posture andoverall treatment which, although, may be explained as a general featureof the Pala art. However, there is no indication to show whether thesculpture belonged to a Causath Yogini complex.

VII. Dancing figure of Vinayaki, set up in Cell No. 49, CausathYogini Temple, at Ranipur-Jhariyal, Bolangir district, Orissa. About 900A.D.^o (ILL 10).

This Yogini figure of Vinayaki is remarkable for several iconographicpoints which are not to be seen in other images of the goddess. Sheis dancing in the caiura pose with her lejs bent to the inside in dance-mo-vements on toes. It is rather a rare dance-pose even in images of Siva,not to say of Gane^a and goddesses.101 Further she is wearing a sarpa-yajnopivlta ( serpen:-is-the sacred-thread ), a typical trait of GaneSa images.She is wearing a sari of which the folds are shown between her t legs andon the sides.

Unfortunately, the two front hands of this four-armed figure aredamaged, although it is presumable from the sharp leftwisc turn of herproboscis, which also is injured in its lower portion, that a bowl of sweetswas originally held in the la^er left hand. The attribute in the upperright hand is not distinct, but the characteristic battle-axe raised in the

100. Cunningham, ASR., vol. XIII, p. 134 ( "22nd. An elephant-headed four-armeidancing female, holding in her one unbroken hand a club." ). Also B. N. Sharma,in Oriental Art, vol. XVIII, No. 3, Autumn 19 72, p. 277, fig. 1.

101. Ibid., p. 280, figs. 2,3tG; C. Sivaramamurti, "Geographical and Chronological Factorsin Indian Iconography," Ancient India, No. 6, Delhi 1950, p. 61; V.S. Agrawala,Siva-Mahadeva, The Great God% Varanasi 1966, pi. XVIII,

26 ] [ The Female Ganesa

extra left hand can be well rec^nis^d. Her ornaments include armlets,bracelets, a torque and a string with -"jewel adorning her high headdressof matted locks, finely arranged in lateral piles from a central parting.As remarked by Sivaramamurti in his study of headgear fashions of theGanesa icons of different areas and epochs, this type of jata-mukuta is afeature characteristic of eastern Indian sculptures of Ganesa as well asSiva.102 Thus its presence in this and the Hirapur YoginI image ofVinayakl is quite in keeping with the regional ioonographic traditions ofOrissa.

VIII. Standing figure of VinayakJ, numbered as YoginI figure 33,Causath YoginI Temple, Hirapur, Bhubaneshwar district, Orissa1 J*(111. 11). The sculpture was described by Mahapatra in the followingwords: "An elephant-faced two-armed figure ( V 9" X 9" ) standingupon the back of an ass and having jata-juta ( knot of matted hair )over her head. Her hands are broken."104 The mistake committed bythis scholar in identifying the mount mouse ( musaka ), on which thegoddess stands, has already been corrected by Sharma.

Unfortunately, all of the four hands of the goddess are brokenand we have no idea as to her attributes. The pot-bellied, heavilymodelled body of the goddess is a beautiful specimen of Orissan artduring the tenth century A. D, She is shown standing elegantly in fron-tal posture, with only slight flexes of tribhanga, on the back of hervehicle mouse. In no other known images of Vinayakl do we find herplaced directly on the mount. In fact, such sculptures of GaneSa orof even Siva are rare which show the deity standing or dancing on hisvehicle.105 The goddess wears only a diaphanous loin-cloth reaching upto the knees and secured by a girdle with its chains hanging betweenthe thighs. Her other ornaments include anklets, armlets, bracelets, amulti-string necklace, and a stringed decorative device on the stylisticjutd-mukuta, quite similar to that seen in the Ranipur-Jhariyal sculpture.

102. Sivaramamurti, op. cit., p. 30, fig. 15, a-c.

103. Kedarnath Mahapatra, UA note on the Hypaethral Temple of Sixty-four Yoginisat Hirapur," Orissa Historical Research Journal, vol. II, No. 2, Bhubaneshwar,July 1953, p. 29; B. N. Sharma, in Oriental Art, vol. XVI, No. 2, p. 169, lig. 5.

104. Mahapatra, op. cit.

105. Getty, Ganela, pi, 7 a\ Sivaramamurti, in Ancient India, No. 6, pi. IV D\ XXV D.

Goddess Vinayakl ] f 27

IX. Image-slab showing four-armed seated figure of Vinayakl,from Hingalajgadh, Mandasor district, Madhya Pradesh, now preser-ved in the State Archaeological Museum, Bhopal. C. 10th centuryA. D. ( ///. 13 ).

Unfortunately, this excellent sculpure is damaged in several of itsportions, including the hands, proboscis and right leg of the goddess.On the elephantine head she is wearing a tiered karanda-mnkuta, adornedwith pearl-strings and a front-jewel. At the back is an elaborate lotus-halo (prabhdcakra ). She is sitting on a lotus-seat with her left legcrossed while the right, which is now broken from the thigh, was per-haps originally suspended. Under her seat a conspicuous figure of rat,her mount, is seen crouching on its hind legs. There are two smallfigures of female attendants standing on either side, both richly dressedand ornamented.

The main figure is clad in a close-fitting diaphanous sari of whichthe folds are finely incised* Her elaborate ornaments include anklets,a girdle, a multi-string pearl-necklace with its strands coming over thebreasts down to the navel and a pendant-feature dangling between thebreasts further below, and armlets. A heavy garland is further remar-kable for its exquisite decoration and the manner in which it is shownhanging through the space between the left thigh and the crossed leg ofthe goddess. The ekadanta trait of her iconography is noticeable butthe trunk below is damaged.

In the upper portion of the slab are carved two miniature femalefigures appearing on lotus-seats flanking the halo of the goddess. Theyare shown sitting in the lahta posture, and from their four-armed natureas well as the abhaya-mudra of right hand they can be well described asgoddesses though their individual identity is not clear. In status theymust be considered as ancillary to the main figure of Vinayaki. The leftside figure is more or less better preserved and the tv\o attributes held inher left hands are a bljapiira and trisula indicating only the generalSaivite context of the representation.

The sculpture presumably belonged to a Causath YoginT group atHingalajgadh, the name of the site indicating the former existence of sucha temple, or of a Tantric goddess temple, there although no architecturalrelics are known except some image-slabs similar to the present piece.

X. Image-slab showing seated figure of Vinayakl, numbered 41, settip in a cell of the Causath Yogini Temple, at Bheraghat, Jabalpur district,

28 ] f The Female Gan^a

NTaJ'iya Pralssh. About 1333 A. D. Inscribed : &n~lingini108 (S;ze117 cm. x 181 cm.) ( ///. 12 ).

This is indeed the most famous figure of the elephant-headed Vini-yaki. She is seen seated on a lotus-leaf nsana with her left leg crossed andright leg pendent on the base below. Unfortunately, many details aremissing and the sculpture is seen in the photograph as assembled fromseveral surviving fragments. Unlike other described images of this goddess,she is seen here as having an attenuated waist and with a figure of anelephant-faced male acting as her support. All her four arms as well as theelephantine trunk are damaged. None of the attributes originally held inher hands can even be guessed now. However, with respect to the ornamen-tation, several details are comparable with the specimen frcm Hirgalaj-gadh described above, particularly the fashicc of the garland treating it aspassing under the bent left leg and encircling the suspended right leg. Asusual with the goddess figures of the period, her upper body is nudeexcept the lavish ornaments, including an elaborate necklace and armlets.A sdri executed with ornate ripple-like folds is seen secured round the waistby a rich girdle with jewelled tassels falling over the thighs.

The back of the settee is embellished with z^a/a-brackets and underthem on either side is seen standing a female attendant carrying a garland.To ihe front on recessed corners of the pedestal are smaller figures ofkneeling devotees, on left being a female and another damaged figurewhile on right a male with folded hands. Below the right knee of thegoddess there appears a kneeling two-armed person with the face of anelephant, presumably Ga^esa, in the pose of supporting the goddess on hisleft hand. This Oanapati portrayal in a subordinate role below the goddessis rather curious and a unique feature of the present sculpture. In theHingalajgadh image the same position under the seat of the goddess isoccupied by the vehicle mouse although there can be no comparisonbetween the particular posture of the male deity here and the simpleanimal figure of the mount. The figure is clad in a lower garment and arich set of ornaments including a girdle with hanging festoons, armlets,pearl-strings in wrists, a necklace and a jewelled string on the ele-phantine head.

The other interesting feature is the presence of a third eye onthe forehead of the goddess which is also found in some of the Ga^esa

106, See above pp. 1-2.

Goddess Vinayakl ] [ 29

images of contemporary times in northern India.107 As observed byGetty, "Unfortunately the trunk is broken on a line with the tusk, butit is long enough to trace its turning to the left/'

The goddess is called by the name ri-Atngint in the inscribed labelon the pedestal. However, the observation made by D. G. Sircar thatthe plthika or pedestal containing the inscription did not belong to thesculpture originally, is incorrect108 and there no doubts can be maintain-ed at least in this respect as the portion of sculpture from the waist ofthe goddess downwards up to the pedestal with its inscription is oneorganic whole. As to the authenticity of this inscribed name if anydoubts were hazarded by some earlier scholars, that was because it is avery obscure name in its meaning. We do not, yet, possess any textualsource which records this curious word for the name of the elephant-headed goddess.

Long ago, Cunningham attempted an explanation of the name'Srl-AingTjLi" as follows : "The name seems to refer to ingga 'movable7,which is itself derived from igi cto go\"109 It is indeed difficult to agreewith this derivation, which, although, cannot be dispensed with as alto-gether impossible. To us it appears that the word Ainginl is a simplePrakrit derivation from Sanskrit compound ''ibhangini1* (ibha=a.n elephant;em_ermf:= having the body ), signifying a female form that has partial body( in this case, the head ) of an elephant. It must be stated that we havenot been able as yet to trace any actual occurrence in texts of the epithetIbhdnginl, or even Ibhdngana in masculine denoting Ganesa, but there aresimilar appellations of the elephant-headed god used by Sanskrit writers,e.g. Ibhlnana, Ibhavaktra,110 consisting the word ibha\ like any other wordsignifying elephant, to denote his elephantine feature. Thus, in an appro-priate manner, the identificatory label inscribed on the pedestal namesthe goddess as ri-Aingini% 'the illustrious elephant-bodied goddess,' thename which was evidently popular with the local tradition in preferenceto other such designations as Vinayaki or GaneSI.

In addition to this Sri-Aingini image, Cunningham described anotherelephant-faced Yogini bearing the name of Jattarl according to the

107. B. N. Sharma, "Ganeia," in A, N.Jha Felicitation Volume, New Delhi 19G9, pp.18 ff.

108. D. C. Sircar, \njAIH., vol. IV, 1970-71, p. 185, fn. 8.109. ASR., vol. IX, p. 68.110. For example, see Vivalocana~koka1 Bo:i bay 1912, p. :>22; Skanda Purani I. 11. 18.

30 ] [ The Female G.mesa

inscription as read by him on its pedestal. As he observed, "38. TattulWilson says a kettle-drum, or any musical instrument. I presume that thename refers to the "trumpet" as the goddess has an elephant's head, andthere is an elephant on the pedestal. Tatta is the imitative sound of thetrumpet, like lanlarara in English." ( ASR., IX, p. 67 ). But when Blochand Banerji visited the place the upper portion of the particular imagehad been already lost. Bloch read its label as Sri-Thathlhari (his No. 73 ),while Banerji and Hira Lai ( No. 43 ) corrected the reading as $ri-Dhadhdhari. The figure is Lumbered by Banerji as 10 and his descriptionruns as follows : "The lower part of a female figure seated on a cushionon the back of an elephant. There is a female kneeling on the left and amale kneeling on the right. The inscription on the pedestal labels heras Sri-DhadhdhariV ( Haihayas of Tripuri, p . 80 ), At the site the imagein question survives to-day only by its lower portion. On the authority ofCunningham,, it is obvious that the same was originally the elephant-headed goddess Vinayak! riding an elephant. The ins:ribed label alsoappears to associate her with D'lundhi, a well-known name of Ganapati,ani the original form of D'vidhdhiri or Dhu[n)dhiri ( which we think is thereading ) can be well recognised as a folk derivation from DJumdhi.

We may recall that i;i some forms of Vajrayana Buddhist images,Ganesa was relegated to a very subordinate position as bcir.g shown tram-pled over by certai.i male and female deities under their feet.111 Therepresentation of elephant-headed god under the feet of his own femaletransformation here in the Yogini sculpture of Bhcraghat is, however,no: of a similar nature showing the sectarian bias on part of an antagoni-stic religious following. In fact, inside the cult of Ganapati himself andpresumably also that of his female aspect Vinayaki, the ancillary beingsor plrsadas and pdrsaut, in attendace upon the main deity were regardedsometimes to have identical elephant-headed aspect and personifiedcharacter bearing similar attributes.112 As has been suggested by Baner-jea,113 the presence of an elephant-headed prostrate figure under thefeet of certain Buddhist divinities might have been inspired from thebelief in the original vighna-c^ciUng trait of the Vinayaka god or gods.

111. Getty, op. cit., pp. 42-5.112. Ibid., p. 12; also see Skanrfa Parana V. I. 70. 32; Bralwianda Pur ana I. IV. 27. 81-

82; \sanakvagurudcvapaihlliati, vol. II, Purvardha XV. 45; XVI. 10-14;Silpm-atna, vol. II, 2:-. IS-49; etc.

i n Din., p 27:..

Goddess Vinayalci ] [ 31

The Nepalese Buddhist deity Vighnantaka and the legend known at theback of its 'obstacle-subduing' character lends some support to such aview for Vighna ataka himself is conceived as assuming the 'obstacle-destroying' role ofGaiiapati himself. But this can only explain one ofthe mythical trends in the process behind the evolution of such newdivinities in later Vajrayana Buddhism, or other pantheons, in additionto the role of sectarian rivalry always present in such iconographic inno-vations where a god is degraded to an ignominious position under thefeet of another divinity.

In case of the Tibetan Mahakala, a representational form was alsoevolved where the figures crushed under his feet included a femaleelephant-faced {being beside the figure of Gaiiea. Some other deitiesshown as standing over the Ganea figure are a Tibetan form of Manjusri,Parnasabari and Aparajita. It is remarkable that a sculpture of the lastnamed goddess was also found from Hingalajgadh ( ///. 16 ) along withthe image of Vinayaki described above from the same complex and isnow deposited in the State Museum, Bhopal. The already known figureof Aparajita from Nalanda presumuly has Buddhist background andcan be well identified as such on the authority of the Sadhanamala,11*but it is difficult to say if we are justified to view the Hingalajgadh speci-men in exclusive Buddhist context. Perhaps, Aparajita was a Yoginigoddess and her mention in the ceremonies presided over by GaneSa,as enumerated in the Mahanirvana-tantra, is of particular significance.116

In 111. 17 we have reproduced a fragmentary sculpture of Aparajitafrom a photograph in the collection of the late Prof. V. S. Agrawala, butthe provenance and present whereabouts of the original are not known.Stylistically the piece belongs to the 9th-10th cent, art of northern Indiaand is quite similar to the Nalanda specimen.

XL Image of bovine-headed goddess Vrsabha, belonging to Yoginicomplex, from Satna, Madhya Pradesh. Now in the Indian Museum,Calcutta. 11th century A. D, On the pedestal the name is inscribed asSri-Vrsabha.11*

114. Bhattacharyya, Indian Buddhist Iconography, 2nd edn, Calcutta 1958, pp. 245-0.

115. See Getty, op. cit., p. 12.

116. ASI, AR., 1925-26, p. 152, pi. LIX; also B. N. Sharma, in Oriental Art, voi,XVIII, No. 3, p. 277, ag. 4.

32 ] [ The Female GaneU

The eight-armed goddess is seated on a lion, and, besides othericonographic details, noteworthy is the figure of baby Ganesa held in thepalm of one of her left hands. Among the votary figures carved on thebase are seen four animal and bird-faced goddesses, those on the Uft ofthe goddess being easily distinguished as representations of pig-headedVisalaksf and elephant-headed Vinayaki. The figure of Vinayaki is seensitting in the lalita posture carrying in her four hands her usual attributes.Her elephantine trunk is turned to the proper left as reaching the bowlof sweets held in the normal left hand. Her form is of the pot-belliedtype as is characteristic in her other Central Indian images. The presenceof the Ganapati figure in the hand of Vrsabha as her child tends to suggestthat this cow-headed Yogini personification was considered to be theprincipal divinity of Saivite nature including in her parivdra both Ganesaand his female equal Vin'iyaki as her children. The relevant myths andiconographic formulations are, however, obscure at present.117

XII. Figure of Vinayakf, in a sculpture on a pillar in theCempakaraman Mandapa, at Suclndram, near Kanyakumari. Of theVijayanagar Period.118

This sculpture of Vinayaki is described by K. K. Pillay as follows :'She is seated in the Sukhasana pose with the right leg kept folded acrossher seat and the other hanging down and resting on a pedestal, ...Herfront right and left hands present the abhaya and varada mudrasrespectively ; her back right hand holds a hatchet, while the corres-ponding left one keeps the pISam. So far, only one other sculpture ofGanesani is known, this is found in a village near Jubbalpore in CentralIndia. The description of that figure tallies word for word withthe Suclndram sculpture.... The only additional element in theCentral Indian sculpture is that, beneath the seat of Gaijesani thereappears a masculine figure with face of an elephant." As corrected byMitra, there is to be seen an ankusa ( i. e. goad ) in the second right handand a pot of modakas in the normal left hand of the goddess.119

117. See Sivaramanmrti, i n / , of the Asiatic Society, vol. XXI, No. 2, 1955, p. 90; same,Early Eastern Calukya Sculpture, Madras 19G2, p. 54.

118. K. K. Pillay, The Sudndram Temple, Adyar 1953, pp. 366-67, pi. 19.

119. Haridas Mitra, Ganapati, p. 116.

Goddess VinayakT ] [ 33

XIII. A late medieval figure of the female form of Ganesa isreported to exist in the Tamresvan Temple at Sadiya in Assam and hasbeen also illustrated.320

XIV. An elephant-headed female in the parivdra of Natesa Siva,depicted in a Pahari miniature of the late 18th cent. A. D.t ChambaSchool, now in the National Museum, New Delhi.121 ( ///. 15 )

The picture shows Siva's twilight dance in which the dancing god isshown surrounded by several divinities and gana figures of his family,inculding GaneSa, Skanda, Parvati besides an elephant-faced femaleon right playing on the drums. B. N. Sharma proposes to identify herwith VinayakT.

XV. Gampati-hrclaya. We have also the Tantric Buddhist goddessname Ganapati-hydnya,122 which is in fact applicable only in a restrictedcontext to a female representation of Ganesa as personifying the Buddhistmantra of that name. As remarked by Getty, the Buddhists claimed"that a mystic mantra in praise of Ganesa, called the Ganapati-hydaya, wasdisclosed to Ananda by the Budcha himself at Rnjagrha. They personifiedthe mantra in the form of a goddess named Ganapati-hrdaya, who, accor-ding to Bhattacharyya, was probably looked upon as the Sakii of Ganesa.The mantra, however, refers only to Ganesa and contains a sadhana to beused in his invocation, when he is to be conceived as being red of hue,standing in a dancing attitude, as having twelve arms holding Tantricsymbols, and as possessed of a third eye as well as of both his tusks."128

The above form as goddess is described in the Dharmakosasamgrahaof Amrtananda in the following words cited by Bhattacharyya withtranslation : "Ganapitihydaya ekamukhd dvibhuja varadd abhayd nYtyasana"Dharmakosasamgraha^ Fol. 43.

"Ganapati-hrdaya is one-faced, two-armed, exhibits in her two handsthe Varada and Abhaya poses, and shows the dancing attitude." Bhatta-charyya also reproduces a miniature representing this goddess in thepossession of Dr W. Y. Evans-Wentz.124 A beautiful painting of this

120. R. M. Nath, The Background of Assamese Culture, Shillong 1948, pi. XXV; D. C.Bbattacharyya, i n / , of the Asiatic Society, vol. VIII, No. 4, p. 265.

121. B. N. Sharma, in Oriental Art, vol. XVIII, No. 3, p. 227, fig. 5.122. B. Bhattacharyya, Indian Buddhist Iconography, 2nd edn, Calcutta 1958, p. 349.123. Getty. Garxeia, p. 37.124. Op. cit., p. 349, Fig, 229.

34 ] [ The Female

goddess is also reported to be contained in an illustrated MSS. of theDh'lranl^amgraha ( No. G. 10741. D ), preserved in the Asiatic Society,Calcutta.125

XVI. A double image of Ganesa is shown by Getty to have beenpopular in China and Japan and presumably based on the Tibetan proto-type of Ganesa in Yab-yum. This formknown in China under thename Kuan-shi-t'ien and in Japan as Kangi-tenconceived two elephant-headed deities, one male and the other female, in embracing posture withtheir elephant-heads posed on the left shoulder of the opposite figure.Particularly in Japan the secret cult of Kangi-ten represented the pairin several other situations.126

125. D. C. Bhattacharyya, in J. of thi Asiatic Society, vol XVIII, No. i} p. 2GG, fn. 1.

126. Getty, op. dt, pp. 72,*80 ff.

APPENDIX I

MULTIPLE GANAPATIS AND THEIR FEMALE SAKTIS

So far we have been able in tracing at least six Lists of fifty or moreGanapatis with their female consorts or Saktis. The earliest of them nodoubt occurs in the Narada Purdna and appears to have been utilised withonly minor variants by the author of the Vighnesddikaldmdtrkd-nyasacontained in the Ucchistaganapati-Updsand. The Nyasa also specifies eachGanesa form and his Sakti to a particular bodily portion and is the onlytext giving this detail.

The lists available are, in fact, of two types. In one case the namesare given with clear-cut mention of the pair of Ganesa and his Saktitogether while in the other two lists, fifcy or more names of Ganapatis aregiven first in one collective group and the names of their fiaktis aregiven next in a similar manner. However, the lists of the second type,as found in the Brahmanda Purdna and the commentary of Raghavabhattaon the Saradatilaka, pose some problems in separating and properlyconnecting the names of the male deity and his female energy togetherin pairs owing to certain doubtful readings and the names being read incertain cases, in dual or plural compounds. Anyway, both types of lists areinteresting and appear to show that they were all derived from one originalset of names,

I. Narada Purdna I. 65. 124-33. Vehktesvara Press, Bombay, edition.

II. Ucchistigan.ipati-Ucchistacandalinyupdsane, pp. 30-34, VenkatevaraPress edn.

III. Jmnirnwatantra, Anandasrama, Poona, edn, 1952, XIV. 61-75.The same is also found quoted by Devanatha Thakkura in hisTantrakaumudi ( completed in A. D. 1554 }, Mithilavidyapitha edn, 1969,Patala HI, pp. 51-52.

IV. Tantrakaumudi of Devanatha Thakkura, Mithilavidyap'tha edn,III, p. 77. It is not known if the author quoted it from some earlier source.

VI. Paddrthddarsa Commentary of Raghavabhatta on the Saraddtilaka( I. 115 ) of Laksmana Desikendra. The same is found quoted in toto byRamatosana Bhattacharya in his digest Prdnatosinl, published byJivananda Vidyasagara, Calcutta 1898, 3rd edn, pp. 83-84.

36 ]

Ndrada Pur ana

Ganesa

1. Vighnesa2. Vighnarfija3. Vinayaka4. Sivottama5. Vighnakrt6. Vighnaharta7. Gananatha

8. Ekadanta

9. Dvidanta10. Gajavaktraka11. Niranjana12. Kaparddf13. Dirghajihva

I( I. 66. 124-38Sakti

HriSriFustiSantiSvastiSarasvatlSvaha

Sumedha

KantiL Kamini

MohiniNatiParvati

14. Sankukarnaka Jvalini

15. Vrsadhvaja16. Ganinayaka17. Gajendra

18. Surpikarna

19. Virocana

20, Lambodara

21. Mahananda

22. Caturmiirti

23. SadaSiva

24. Amoda

25. Durmukha

26. Sumukha

NandaSuresTKamaru-piniUma

Tejovati

Satya

VighnesI

SvarupinI

Kamada

Madajihva

BhQtl

Bhautiki

[ The

11

Female Gaue^a

) UcchistagGnapatyupasana ( 11 )

Ganesa

VighnesaVighnara jaVin'iyakaSivottamaVighnakrtVighnahartaGana

Ekadanta

DvidantaGajavaktraNiranjanaKapardiDirghajihva

Saiikukarna

Sakti

HriSriPustiSantiSvastiSarasvatlSvaha

Sumeiha

KantiKaminiMohiniNatiParvati

Jvalini

Vrsabhadhvaja NandaGanesaGajendra

Surpakarna

Trilocana

Lambodara

Mahananda

Caturmurti

Sada^iva

Amoda

Durmukha

Sumukha

Suresi

Body-portion

LalataMukhavrttaDaksi^anetraVamanetraDaksinakarnaVamakar^aDaksinana-

sa'putaVainana-saputaDaksinagandaVamagandaUrdhvosthaAdharosthaOrdhvadan-

tapafiktiAdhodanta-

panktiSirasMukha

Kamarupial Daksi^aba-

Uma

Tejovati

Satya

Vighne4i

SvarupinI

Kamada

humulaDaksina-

lcfrrnara

Daksina-magibandhaDaksan-

gulimulaDaksan-

giilyagraVamaba-

humulaVamaku-i

rparaManajihva Vamamani-

Bhutl

Bhautika

bandhaVarna ngu-

Vamangu-lyagra

Goddess Viniyaki ]

I

[37

IINdrada Pur ova

27. Pramoda

23. Ekapada29. Dvijihva30. Sura

3i. Vira

32. Santnukha

33. Varada34. Vamadevesa35. Vakratunda

36. Dviranda

37. Senani

38. Kamandha39. Matta40. Vimatta41. Mattavaha42. Jati43. Mundi44. Khadgi45. Varenya46. Vrsaketana

47. Bhaksya-priya

48, Gaijesa

(I . 66. 124-38

Sita

RamaMahisiJambhinI

Vikar^a

BhrkutI

LajjaDirghaghonaDhanur-dhariYaminf

Ratri

GramanlSasiprabhaLolanetraCancalaDiptiSubhagaDurbhagaSiva

L Bhaga

Bhagini

Bhagini(Bhoginl)

49. Meghanada Subhaga

50. Vyapi

51. Games'vara

(Subhaga)1

Kalaratri

, Kalika

) Ucc hist aga n apa tyupdsan a

Pramoda

EkapadaDvijihva6ura

Vira

anmukha

VaradaVamadevaVakratunda

Dvirada

Senanl

KamandhaMattaVimattaMattavahanaJatiMundiKhadgiVarenyaVrsaketana

Bhaktipriya

Ganesa

Meghanada

Vyapti(Vyapf)*Ganesvara

Sita

RamaMahisiBhanjani

Vikarana

BhrkutI

Lajja

( I I )Daksapada-

mulaDaksin^januDaksinagulphaDaksinapada-

rigulimulaDaksinapa-

dangulyagraVamapa-

damulaVamajanu

Dirghaghona VamagulphaDhanurdhara Vamapa-

Yamini

Ratri

GramaniSa^iprabhaLolalocanaCancaliDiptiSubhagaDurbhagaSivaBhaga

Bhagini

Bhoginl

Subhaga(Subhaga)Kalaratri

Kalika

dahgulimulaVamapada-iigu.ydgra

Diksiua-parsva

VamaparsvaPrsthaNabhiJatharaHrd

DaksamsaKakud*VamamsaHrdayadida-IcsahastantaHrdayadi-

vamahastantaHrdayadi-

daksapadantaHrdayidiva-mapadantaJathara

Mukha

1. In view of the fact that the name Bhagini is read above and the reading in List IIis Bhoginl in this position, we may emend it as Bhogini. In both the Lists thename Subhaga may be, similarly, emended as Subhaga for SubJiaga is read aboveas No. 43.

3. In view of the rexdmgjfyapl in List I and elsewhere, Vyapti may be taken as acorrupt text for VyapJ.

38 ] [ The Female Gaacs.i

IIIJndnarnava ( XIV.

1. Vighnesvara2. Vighnaraja3. Vinayaka4. Sivottama5. Vighnakrt5

6. Vighnahrt6

7. Vighnarat8, Gananayaka9. Ekadanta

10. Dvidauta11. G^javaktra12. Niranjana13. Kapardi7

14. Dirghavaktra15, Sahkarsana16. Vr?adhvaja17. Gananatha18. Gajendra19. Surpakarna10

20. Trinetra21. Lambodara

IV61-753 ) Tantrakaumudi of Devanatha (III, p. 77 )

Sri

Hrl4

TustiSantiPustiSarasvatrRatiMedhaKantiKaminIMohiniJat'iTivraJvaliniNandaubhaga8

VighnesaVighnarajaVinayakaSivottamaVighnahrtVighnakartaVighnaratGananayakaEkadantaDvidantaGajavaktraNiranjanaKapardiDirghavaktraSaiikarsanaVrsadhvaja

Kamarupinika GanaaathaSubhra9

JayinISatyaVighnesi

GajendraSurpakarnaTrilocanaLambodara

SriKriyaTustiSantiPustiSarasvatIRatiMedhaKamiKaminIMohiniJataTivraJvaliniNandaSurasaKamarupinlgubhraJayinISatyaVighnesi

3. The printed text of Anandasr.ima, Poona, gives variant readings of three MSS.wliich are noted below wherever thought significant. The variants as preserved inthe Tantrakaumudi citation of this passage are also of interest and are given belowby mentioning Devanatha.

4. Devanatha .' Kriya : which is evidently a corrupt reading for lhriya* of thePoona edn.

5. Devanatha .' Vighnahrt.

6. Devanatha : Vighnakrt.

7. Devanatha .' Kapardablirt.

8. Devanatha : Surasa : which is also noted as variant text in the Poona edn.

9. Devanatha .* Subha.

10. Devanatha : Surpakarna I Poona edn. variant Suryyako(i.

Ccddcss Vinlyaki ] [ 39

IIIJndnnrnava (

22* Mahanada11

23. Caturmurti12

24. Sadasiva25. Amoda26. Durrnukha27. Sumukha28. Pramoda29. Ekapada30. Dvijihva31. Sura19

32. Vlra33. Sanmukha34. Varada35. Vamadeva36. Vakratunda37. Dvirandaka38. Senani39. Gramani22

40. Matta41. Vimatta42. Mattavahana43. JatI

IVXIV. 61-75 ) Tantrakaumudi of Devanatha (III. p. 77 )

SvariipinlKamadaMadavihvalaVikataGhuriia13

Bhuti1*BhumiSatiRamaManusTMakaradhvajaVikarna20

BhrkutiLajjaDfrghaghoijaDhanurdharaYarnin!21

RatriCa^dika23

SasiprabhaLolaCapaleksa^fx

Mahananda15

Candamurtl16

SadasivaAmodaDurmukhaSumukhaPramodaEkapadaDvijihvaSuraViraSanmukhaVaradaVamadevaVakratundaDvirandakaSenaniGraminiTamasVimalaMattavahanaJati

SvariipinlKamadaMadavihvalaVikataPancaslrsa17

Purnima18

BhumiSatiRama.Manus!MakaradhvajiVivarnaBhrukutiLajjaDirghaghonaDhanurdharaMalinIRatriCandrik aSasiprabhaLolaCapaleksana

11. Devanatha : MahZnanda.12. Devanatha ; Candamurti'.13. Devanatha : Purna.14. Devanatha : Bhutima, which is evidently n corrupt reading for %bUutim'dtit of the

Poona edn.

15. In this Tantfctliaumudi passage the author Devanathn is evidently citing from so.ncearlier source and also gives variants in four cases as follows : Mahan'ad'i*

16. Caturmurti,17. Puina.18. Bhuti.19. Devanatha .' Sura,20. Devanatha : Vivarna.21. Devanatha : MalinI.22. Devanatha .* Gramini,23. Devanatha ;

40 J [ The Female Ganesa

IIIJndnarnava (

44. Mundi45- Khadgi25

46. Varenya57. Vrsaketu48. Bhaksyapriya49. Meghanada50. Ganesa28

51. Ganapa29

XIV. 61-75)

Rjvi24

Durbhaga26

SubhagaSivaDurgaKalik\27

Kalakubjl

IVTantrakaumudi of Devanatha (HI p. 11)

MundiSadvinduVarenyaVrsaketuBhaksyapriya

MeghanadaGajesa

Vighnaharini30 Ganesa

RuksaBhagaSubhagaSivaDurgaKaminiKalakubjaVisvahariai

^4. Devanatha i Rukfd : Poona cdituin variant

25. Devanatha .' Sadvitulu,

26. Devanatha .' Bhaga.

27. Devanatha : Kaminl.

28. Devan a tha : GajeL.

20. Devanatha .' G(fnchr4

SO, Devanatha ; Viivahan'nl,

Goddess .Vinayafci ] [41

Brahmdndi Purani( II. IV. 44. 63-76 )

VI

PadarlhdlarSa Comm. of Raghava-bhatta on the Sdraddtilaka ( I . 115 )

of Laksmanadesikendra

1. Vighnesa2. Vighnaraja

3. Vinslyaka

4. Sivottama5. Vighnakrt

6. Vighnahanta7. Vighnarat

8. Gananayaka

9. Ekadanta

10. Dvidanta

11. Gajavaktra

12. Niranjana

13. Kapardavan

14. Dirgharnukha

15. Sankukarna

16. Vfsadhvaja

17. Gananatha

18. Gajendrasya

19. Surpakarna20. Trilocana21. Lambodara

22. Mahanada

SriHri

Pusti

SantiTusti

SarasvatiRati

Medh'i

Kanti

KaminI

Mohini

( Jata )32

TivraJvalini

Nanda

Suyasah

Kamarupim

Ugra

Tejovati

Satya

Vighnesani

Svarupini

VighnesaVighnaraja

Vinayaka

Sivottama

Vighnakrt34

Vighnaharta36

Gaii a

Ekadanta

Dvidanta

Sudanta37

GajjvaktraNiranjana

KapardJ

Dirghajihvaka

Sankukarna

Vrsabhadhvaja

Gaaanayaka

Gajendra

Surpakarna11

TrilocanaLambodaraMahananda

Hrlrf

Pusti

SaniiSvasti35

SarasvatiSvaha

Medha

Kanti

KaminI

Mohini

Nati

Parvati

Jvalini88

Nanda

Supas"a3fl

KamarupiniUma40

TejovatT

SatyaVighnesani

Surupin!

31. Lists V and VI give nimes of Gines^is uid the aktis in two separate groups.32. The. Venkatesvara Press edn, used here, reads Hatlia\ which appears to be the

corrupt reading of original Jata or NaH, as found in other Lists.The same is cited by R:\mat03i~ni Bha^acharya in his compilation Pt-anatopni^which is at variance in case of certain readings in the printed edn. of the Sarada-Hlaka Commentnry. All such significant variants'are noted below.P/3fl#. : Vtghnahrt.Ibid. / Ksanti.Ibid. : Vighnakartta.The original has : 'ganzikadvisudantakah', which is to be split as giving fournames of Gana, Ekadanta, Dvidanta and Sudanta.

38. Pratia. : Jvalini.89, Ibid. ' Suyaiafi. Cp. Sureli elsewhere.40# Ibid. : Ugra. Note that 'ugratejovatl* is one word in the printed text of the

Praixa.41. Ibid. ; Suryyakartict. .

33.

34.35.36.37.

^ JV

Brahmdyfa

23. Caturmurti24. SadaSiva25. Amoda26. Durmada27. Sumukha28. Pramodaka29. Ekapada30. Dvipada31. Sura32. VJra33. Saamukha34. Varada

Purayd

KamartaMadajihvaVikataGhtirnitananaBhutiBhumiDviramyaAmarupaMakaradhvajaVikar^aBhrukutiLajja

35. Nama(Vama)deva Dirghaghona36. Vakratunda37. Dvidantaka38. Senanf39. Gramaijf40. Matta41. Mattamusa-

kavahana42. Jati43. Muijdi44. Khadgi45. Varenya46. Vrsaketana47. Bhaksyapriya48. Gaqea49. Mcghanada50. GaiieSvara

42. Prana. ; Amogha.44. Ibid. : Ekap-ada*45. Ibid. .* Sivamukha .'46. Ibid..' Subhaga.48. Ibid. . Garbha.

DhanurdhariYaminiRatriCandrakantaSasiprabhaLolaksi

CapalaRjviDurbhaga

SubhagaSivaDurgaGuhapriyaKaliKalajihva

t The Female Gage3;

VIComm. on the

CaturmurtiSadasivaAmoda42

DurmukhaSumukhaPramodakaEkarada44

DvijihvakaSuraViraSanmukha45

VaradaVamadevaVakratu^daDviraudakaSenaniGramaniMattaVimatta

MattavahanaJati

Mundl

KhadgiVarenyaVrsaketanaBhaksapriya49

Gane^aMeghanadaka50

VyaplGanesvara

43. It.may be even -

Sdradatilaka

KamadaMadajihvaBhutiBhautikaSita48

RamaMahisiBhanjin!Vikar^apaBhrukutiLajjaDfrghaghoDaDhanurdharaYaminiRatriSamjria

KamandhaSasiprabhciLolaksi

CancalaDfpti

Durbhaga40

Subhaga47

SivaBharga48

BhaginiBhoginiSubhagaKalaratriKalika"

Asitt.

the reading here being 'sHrciviraiivamukhahs47. Ibid, : Durbhaga.49. Ibid. ; Bhak^yapriya.

50. Ibid. ; Meghanayaka.51. Thus it comes to have 52 pairs of names.

WOttDlNDEX

Acdrodinakara 13Agni-Purdna 8Aingini 12, 2830AlaksmI 67, 8Alaksmi-ndsana 7Amaravati 4Apirajita 3IArdhanfiri Gayea 1012 156Ardhanarisvara 11Aryamanjusrimulakalpa 8> 12, 19, 22Asta-matrka 14, 17, 21-22Astasiddhi 17Atharvaveda 6

Bajanath Prasasti 15Bali 15Bdlidvipa grant hah 145Baudhdyana Dharmasutra 6Bhairavapadmdvatlkalpa 13Bhavisya Pur ana 17Bheraghata 1, 17, 2730Bhumara 18Bodhdyana Grhyapariiitas 5,7, 8-9, 20Brahma 17Brahmai^aspati 6> 7Brahmani 22Brahml 17Brahmdnda Purdna 10,14, 16, 35, 41-2Buddhi 17Buddhist icons 30-31, 33-4Gausath Yogint see YoginICamu^da 22Danti 5, 6Dharmakosasamgraha 33Devagarh 4phuijdhi 30Elephant-faced Malinf 15

Gadhwa 14Gajamukha 9, 19Gajamukhl 9Gajanana 3, 13, 19Gajasirsa Yaksa 4Gajasya 9Gajavaktra 9, 19Ga^apati see GanesaGanapatKhrdaya. 2f 23-4Ganendra 15Ganendri 14-5Ganesa 1 ff.Ganesa-maivka 10Gane^ani 13, 19Ga^esi 3, 19G a ^ v a n 3 t 13, 19Giriyek 245Guhyasamdjatanlra 12Halayudha Stotra 15ffaracorita 15Harivatnsa 9Harshagiri 234Hastimukha 56Hastimukha 7, 8, 19, 20

- Jyestha see JyethaHayagrlvi ISHingaUjgadh 6 27S, 31Hirapur 26Indra 17IndraijI 145, 17, 22Isdnai ivagurudevapaddhat i 14-16Jnmarnavatanira 10, 35, 3840Jyestha 6

- raja 6Jyestha 5,78,20

- Kalpa 57

44 ]

Kangi-ten 34Kaumari 178, 22Kuan-shi-t'ien 34Kumara 17 8Kumar! 14 ,LaksmT 6, 7, 18Lalitdsahasrandma 10Lambodarl 14, 19Unga Pur ana 8Mahakala 31Mahalaksmi 7Mahdnirvdnatantra 31Mahesvari 14-5, 22Maitrdyani Samhitd 5Malinl, elephant-faced 15Mcinava Grhyasutra 5Manjusrl 31Mathura 4-5, 9, 20-1Nalanda 31Mrada Purdna \03 35, 36-7 .Narasimha 17, 18Narasimhl 14, 17-8, 21Paruasabarl 31Prdnatosinl 35, 41-2Rairfa 5, 9, 20-1Ranipur-Jhariyal 25, 26 .Rddhil7, 18Rgveda 6Sadiya 33Sakti-Ganapati 10-11, 15-6, 18Sakti-Ganesvara 10Sdradatilaka 10, 35, 41-2Siddhi 17, 18Silparatna 10-12, 15-6

[ Word-Index

Siva 8, )2, 14,33Sixtyfour Yoginls, see YoginISkanda Purdna 7, 9, 14, 16, 22Suhania 23Sri-Aingini, see Aingmi$rJ-Dhadhdhari 30SucTndram temple 32Tcitliriya Arcvyaka 5Tamresvarl temple 33Tantrakaumudi 10,35,3840

Tattari 29UcchistaganaJ)atyupdsand 10, 357Udaigiri 4VainayakI, see VinayakIVaisnavi 145, 18, 22Vanadur gopanisad 13,22Varaha 17, 18Varahi 178,22Varunanl 14, 21Vdyu Purdna 9Vedaparikrama 14Vidhimdrgaprapd 13Vighnantaka 31 ;Vighnesa-mdtrka 10Vighnesam 10Vighne^i3, 10-19Vinayaka see GanesaVinayaka 3. 8, 12VinayakI 1 ff.Vi^alaksi 32Visnul8Visnudharmottara Purdna 8Vrsabha31 2 -YoginI 12,124, 22 ff.