¹ The Hindu lunar month is conceived of as the interval between two consecutive oppositions or two consecutive conjunctions of the sun and moon. The pūurṇimā system followed in North India measures the month from pūrṇīmā (full moon or opposition) to pūrṇīmā; the amānta system followed in Maharashtra and most of South India measures it from amāvasyā (no moon or conjunction) to amāvasyā. SeeUnderhill, M. M., The Hindu Religious Year (Calcutta: Association Press, 1921), p. 23Google Scholar and Somayaji, D. A., A Critical Study of the Ancient Hindu Astronomy (Dharwar: Registrar Karnatak Univ., 1971), ch. 3, Nos. 2.5, 3.1, 3.3.Google Scholar

² I am here using the term “solar day” to translate the Marathi word vār (day of the week, usually measured from sunrise to sunrise). Somayaji (n. 1 above), in ch. 3, No. 2.1, translates this term “civil day,” reserving the term “solar day” for saurāha (1/360 of a solar year—measured from summer solstice to summer solstice).

³ I tithi = 60 ghaṭis; 1 ghaṭi (ca. 24 minutes) = 60 vighaṭis; 1 vighaṭi = 60 paraghaṭis; 1 paraghaṭi = 60 sukṣma ghaṭis. See Somayaji (n. 1 above), ch. 3, No. 2.6.

⁴ Thus, for example, strictly speaking, the names caturthīand ekādaśī refer to the fourth and eleventh tithis of any lunar month, but the rituals and fasts associated with these tithis are observed by nearly all Hindus throughout the whole of the solar day (vār) to which they correspond; similarly with the tithis amāvasyā and pūrṇīmā. The naming of the solar days, however, is not precisely sequential. Since tithis vary from approximately 21.5 to 26 hours, occasionally (about 13 times per year) a tithi will expire between the sunrises of successive solar days. When this happens, the name of that tithi is normally omitted in the almanac. Similarly, when two solar days begin during the same tithi, a double solar day is normally projected. It should be noted, however, that there is more than one way of determining the precise correspondence of tithi and vār; astrologers and almanacs are not always in agreement.

⁵ Bhat, M. R. (Modern Astrology, Delhi: Motilal Banarsidass, 1967, pp. 47 ff.)Google Scholar suggests that the amāvasyā tithi should be divided into eight equal parts. The second through the fifth are moderately inauspicious; the fifth, very inauspicious; and the last two parts, most inauspicious, “the evil effects… untold and manifold.” Amāvasyā is the one day in the month in which the gargācārya muhūtra, the usually very auspicious time just before dawn, is inauspicious.

⁶ Information regarding popular beliefs about amāvasyā is based on interviews of Maharashtrian informants and reports from British ethnographers: inter alia, Abbott, J., The Keys of Power (London: Methuen, 1932)Google Scholar; Crooke, William, The Popular Religion and Folklore of Northern India (Delhi: Munshiram Manoharlal, rev. ed. 1968, 2 vols.)Google Scholar and Religion and Folklore of Northern India [hereafter R & F] (London: Oxford Univ. Press, 1926)Google Scholar; Russell, R. V., The Tribes and Castes of the Central Provinces, London: Macmillan, 1916, 4 vols.)Google Scholar; and Enrhoven, R. E., The Folklore of Bombay (Oxford: Clarendon Press, 1924).Google Scholar Except where noted, the views reported by ethnographers have been confirmed by interviews and observations.

⁷ Abbott (n. 6 above), p. 335; cf. p. 368.

⁸ See Enthoven (n. 6 above), p. 57; confirmed in interviews.

⁹ Abbott (n. 6 above), p. 259; confirmed in interviews.

¹⁰ Abbott, p. 260. Present-day villagers, however, generally consider the animals too valuable to give away. They prefer to run the risk, trusting the power of pūjās to avert the bad effects of a malefic birth.

¹¹ See Crooke, Popular (n. 6 above), vol. II, p. 234; confirmed in interviews. The exception to this rule is that bullocks used to transport pilgrims and images to Jejuri on Somavati Amāvasyā are expected to gain power from the amāvasyā work.

¹² See Abbott (n. 6 above), pp. 178, 230–31; confirmed in interviews. Amāvasyā pūjās are also performed by owners of trucks, rickshaws, bicycles, handcarts, etc.

¹³ See Abbott, p. 54.

¹⁴ Ibid., p. 335; confirmed in interviews.

¹⁵ Abbott, p. 59; confirmed in interviews.

¹⁶ Underhill (n. 1 above), p. 36; confirmed by observation.

¹⁷ See note 19 below.

¹⁸ See Abbott (n. 6 above) p. 268. Explicit connections between amāvasyā and eclipse are rare; implicit connections, however, abound. Most of the same restrictions and recommended observances apply to both times (see Abbott, pp. 488, 264ff.;O'Malley, L. S. S., Popular Hinduism: The Religion of the Masses [Cambridge: University Press, 1935], p. 124Google Scholar; and Underhill [n. 1 above], pp. 67 ff). Demons and ghosts are rampant at both times (see Crooke, R & F, p. 40). Amāvasyā is popularly conceived of both as the death of the moon and as sexual intercourse between the sun and the moon. Similarly, many of the myths and legends explaining the eclipse of the moon associate it in some way with a moment of pollution, e.g., (i) sexual intercourse between the sun and moon; (ii) the shadow of a low-caste person (Māṅg, Mahār, or Bhaṅgi) falling across the moon (Modi, J. J., “An-cient Beliefs about Eclipse, ”Journal of the Anthropological Society of Bombay, III, 6 [1916], p. 359)Google Scholar; (iii) the indebtedness of the moon or the gods to a low-caste person—Māng or Chāmbhār (R. K. Battacharyya, “A Note on Some Curious Superstitions about the Eclipse of the Moon,” ibid., XIII, 4 [1925], pp. 363ff.); (iv) the swallowing of the moon by an unclean animal such as a toad or dog (Crooke, R&F, p. 39) or the demon Rahu. The giving of alms to outcastes on amāvasyā closely parallels the practice at an eclipse. The demon Rahu is both worshiped (Enthoven [n. 6 above], p. 60; cf. Crooke, R'F, p. 40) and driven away by noise (Broughton, T. D., The Costume, Character, Manners, Domestic Habits, and Religious Ceremonies of the Marhattas [London: John Murray, 1813], pp. 14fGoogle Scholar during an eclipse; during certain amāvasyās (Āśvina, Bhādrapada, and Śrāvana), the power of Rahu is invoked into darbha grass. Dogs, sometimes chased with sticks at the time of an eclipse, are worshiped in Jejuri on Somavati Amāvasyā.

¹⁹ Abbott (n. 6 above), p. 251; cf. Gupte, R. B., Hindu Holidays and Ceremonies (Calcutta: Thacker, Spink & Co., 1919), pp. 30–32Google Scholar. Parva = knot or tying together; kāl = a time or moment. A parvakāl is a moment of juncture or coming together, resulting in power. It is not to be confused with a muhūrta, which is a carefully calculated, unambivalently auspicious moment (see Raman, B. V., Hindu Predictive Astrology [Gandhinagar: I. B. H. Prakashana, 1972], pp. 274ff.).Google Scholar A parvakāl enhances all power—malefic as well as benefic. It is noteworthy that, in popular Hindu lore, nearly any place or moment of juncture has power (śakti): a place where roads or paths meet has power (a meeting of three roads has the special power to cure ghost possession); two tree trunks that have grown together have sacred power; saṅgams (confluences of rivers) have a sacred power nearly always marked by a temple; the point where two streams of pilgrims meet on a pilgrimage has power, and the moment of joining is marked by a ritual (e.g., in the Paṇḍharpur pilgrimage, bhajan singing at the meeting of dindis; in the Jejuri pilgrimage, shouts of “Sadānandācā Yelkot”). Similarly, the moment of conjunction of any two celestial bodies is a moment of power, a śakti time, a parvakāl.

²⁰ Some believe this is especially true on Soma-vati Amāvasyā. For example, Rāmoshis who take an oath on a Somavati Amāvasyā while in contact with Khandobā's bhandār (yellow turmeric powder) are bound to the oath to their death (Gazetteer of Bombay State, XX, rev. ed., District Series, 1954, p. 126).

²¹ Abbott (n. 6 above), pp. 204f.

²² See ibid., pp. 166–67. This may well reflect the belief that the power of the moon has entered the waters.

²³ Ibid., p. 260; confirmed in interviews.

²⁴ Underhill (n. 1 above), p. 42; cf. Abbott (n. 6 above), p. 441.

²⁵ This is evident in the cultic activity associated with the navagraha. The pūjāris at Navagraha Mandir, near Shanwar Wada, Poona, claim that more than twice as many people come to that temple on an amāvasyā as on a normal day. On a śani amāvasyā (an amāvasyā falling on a Saturday), the god Śani (=Saturn) is especially open to peti-tions to have his 7–1/2 years of evil influence (sāḍesāti) mitigated, and to grant peace (śānti) to those who take darśan and make pūjā. In the vil-lage Shani Shingnapur in Ahmednagar District, there is a large yātrā on every śani amāvasyā because of the heightened fluidity of Sani's power on that day (taped interview, Śani Amāvasyā, 10 May 1975)

²⁶ Buddhavār (Wednesday) is governed by Bud-dha (Mercury); Guruvār (Thursday) by Guru or Bṛhaspati (Jupiter); Śukravār (Friday) by Śukra, literally “semen” (Venus); Śanivar (Saturday) by Śani (Saturn).

²⁷ Enthoven (n. 6 above), p. 54; Crooke, R'F, p. 407.

²⁸ Abbott (n. 6 above), pp. 182–83.

²⁹ In addition to Khaṇḍobā, the most important of these are Mhaskobā, Jyotiba, and Bhairav. Eight separate forms of Bhairav are usually distin-guished, of which one is Mārtaṇḍ Bhairav; most believe that Mārtaṇḍ Bhairav is Khaṇḍobā, though in the most sophisticated mythology Khaṇḍoba is believed to be an avatār of Mārtaṇḍ Bhairav. Many Maharashtrians believe that Jyotiba and Mhaskobā are also avatārs of Khaṇḍobā.

³⁰ There are over 600 Khaṇḍobā temples in the Deccan. Eleven of these (6 in Maharashtra and 5 in northern Karnatak) are considered the especially powerful (jāgrit) places of Khaṇḍobā. Though four in addition to Jejuri claim to be Khaṇḍobā's original place, Jejuri's claim is most widely accepted and is undisputed in the northern cult.

³¹ A navas is a vow made to Khaṇḍobā to perform some service or extraordinary fear for him in return for which he grants a request. The requests are usually simple and specific (e.g., good crops, children, victory in a wrestling match, money, success in a legal trial), though sometimes more general (happiness, success, wealth, etc.). The vows are usually to fulfill simple services to the god (to make a pilgrimage, sacrifice a goat, make pūjā a certain number of times, etc.)-Sometimes (but rarely in present times) more extreme vows are made—e.g., dedication of a child. In the recent past, extreme vows involving extraordinary feats of strength and endurance of pain (such as hook-swinging and walking over hot coals) were common at Jejuri.

³² Russell (n. 6 above), II, p. 43.

³³ Turmeric is a multivocal symbol appearing in many different contexts in popular Hinduism; for a treatment of it as a sun symbol outside the context of the Khaṇḍobā cult, see Dymock, W., “On the Use of Turmeric in Hindu Ceremonial,” J. Anthro. Soc. of Bombay, XI, 7 (1891), pp. 445–46.Google Scholar

³⁴ A late Sanskrit work, alleged to be the fifth part of the Brhmandā Purāna. I have worked with two different English versions: one, a literal English translation from the Sanskrit, done for me in 1970 by P. N. Navathe of Poona; the other, made available to me by McKim Marriott, a hand-written English translation (by an unnamed trans lator) from the Marathi version of Siddhapal Ke-sari.

³⁵ Malhāri Mahātmyā 8:3–8. Cf. 1:40, 4:1–5; 4:41; 6:12; 7:3.

³⁶ Extensive interviews at 9 of Khaṇḍobā's jāgrit temples and at 50 temples of lesser importance indicate that Jejuri is the undisputed center of the northern portion of the cult; Mailār in Be-lari, Dewargudda ⁱⁿ Dharwar, and Mangsuli in Belgaum are rival centers in the South. No Soma-vati Amāvasyā pilgrimages are made to any of the southern temples, but some devotees from the south make regular pilgrimages to Jejuri.

³⁷ (i) Caitra Pūrnimā celebrates Śiva's incar-nation into the Khaṇḍobāavatār, (2) Pauṣa Pūrnimā celebrates the marriage of Khandobā to Mhāl-sābāi, (3) Māgha Pūrnimā is the observance of Mhālsābāi's birthday, (4) Āśvina Vijayādaśamī (Dasarā) observes the visit of Khaṇḍobā of Gad-kot to the Khaṇḍobā of Karhe-pathār, and (5) Mār-gaśīrṣa Campāṣaṣthī observes the victory of Khaṇḍobā over the demons Mani and Malla.

³⁸ Estimates of attendance at festivals vary a great deal. Because pilgrim tax was not collected at Somavati Amāvasyā festival before June 1975, officials at Jejuri do not have precise figures. All agree, however, that Somavati Amāvasyā festivals are usually much larger than any others, though Campāṣaṣthī sometimes is almost as large. In 1961, the Bureau of Economics and Statistics of the Govt, of Maharashtra kept statistics, by village and taluka, on fairs and festivals. According to their figures (which are conservative), about 25,000 people attended the Dasarā Yātrā only 2,000 attended each of the other four festivals (Socio Economic Review and District Statistical Abstract, Poona District [Poona: Govt, of Maharashtra, 1961–62], p. 104). The 1961 Census of India gives figures only for the Campāṣaṣthī festival, estimating attendance at 30,000–40,000. Somavati Amāvasyā attendance is not recorded in either study, since there is no fair associated with it; but Mate, M. S. (Temples and Legends of Maharashtra, Bombay: Bharatiya Vidya Bhavan, 1962, p. 173) reports attendance at the 1961 Somavati Amāvasyā festivals as 70,000 (Śrāvana) and 60,000 (Pauṣa). The double Somavati Amāvasyā which I observed (9–11 Feb 1975) was attended by over 100,000 pilgrims.Google Scholar

³⁹ An exception to this occurs when the amāvasyā tithi encompasses the sunrises of two solar days (Sunday and Monday, or Monday and Tuesday); then the yātrā can be declared a two-day (Sunday and Monday) or three-day (Sunday, Mon-day, and Tuesday) event. In these cases, entertain-ment facilities are available (tamāśā tents, etc.), but no major trading takes place.

⁴⁰ Mate (n. 38 above);Dhere, R. C., Khandobā (Poona: R.J. Deshmukh & Co., 1961)Google Scholar; Khare, G. H., Mahārāṣtraci Car Daivate (Poona: G. A. Khare Prakashak, 1958) and “Khandobā-Malhāri-Mailār: Ek Lokdeo,” Kesari (29 Dec 1957)Google Scholar; Babar, Sarojini (ed.) Kuldaivat (Bombay: Mahārāṣtra Rājya Loksāhitya Samiti, 1974).Google Scholar The Mārtand Vijay, a Marathi grantha, exists only in a manuscript form.

⁴¹ The following description is based on observations at, and participation in, five different Somavati Amāvasyā festivals at Jejuri (31 Aug and 28 Dec 1970; to Feb, 9 Jun, and 3 Nov 1975), and interviews with pilgrims, wāghyās, village officials, and elders at Jejuri (Aug-Mar 1970 and Jan-Jun 1975).

⁴² An open palanquin used to carry the image of a god in procession. In Jejuri the pālakhi is exceptionally large, about 6 feet long and 4 feet wide, with a carrying pole 6 to 8 inches in diameter and about 18 feet long. Made of extremely dense wood, it weighs over 300 pounds.

⁴³ There are two temples at Jejuri: Karhe-pathār sits high above the village; Gad-kot is also on a hill, but much nearer the village. Karhe-pathār is considered the older and generally more important of the two temples. On most festivals, including Somavati Amāvasyā, the most serious pilgrims will visit both temples; but there is a long tradition, supported by a charter myth for the existence of the temple at Gaḍ-kot, that a visit to Gaḍ-kot. suffices.

⁴⁴ These rhythmic ritual slogans are chanted an-tiphonally at important moments at all Khaṇḍobā festivals. Sadānandācā is a possessive form, literally “of the everlasting bliss”; “Yelkot” and “Yelkāri” are epithets of Khaṇḍobā.

⁴⁵ For over ten years, Babulal Rasul Khan, a retired police officer in Jejuri, has had the honor of leading Khaṇḍobā's horse in the procession. A Muslim, Rasul Khan considers Allah his family god (kuldaivat), but regards Khaṇḍobā as his chosen god (iṣṭadaivat).

⁴⁶ Dogs used to be a part of the procession in Jejuri. In most cases now, however, their owners keep them some distance from the actual procession for fear they will be trampled.

⁴⁷ The laṅgar does not always break; usually the links of the chain separate. Still, great force is required to accomplish even this. The laṅgar itself is considered sacred, has a number of other ritual functions in the cult, and is regarded as an avatār of Śiva's snake.

⁴⁸ Sachau, E. C. (trans.),Alberuni'sIndia (London: Trübner, 1888), II, p. 185; cf. I, p. 28.Google Scholar

⁴⁹ The Kali Yuga is the last and the Krta Yuga the first of the four yugas (ages) in Hindu chronological theory. At the end of the Kali Yuga, the cycle of yugas (i.e., one æon or mahāyuga) is finished; the æon is reabsorbed (or inhaled) into Brahmā's being, and another æon beginning with the Krta Yuga emanates (or is breathed out). See Basak, R., “The Hindu Concept of the Natu-ral World” in Morgan, K. W. (ed.), The Religion of the Hindus (New York: Ronald Press, 1953), p. 89Google Scholar and Danielou, A., Hindu Polytheism (London: Routledge and Kegan Paul, 1964), p. 249.Google Scholar The Kali Yuga lasts 432,000 human years; the other three yugas last two, three, and four times that figure, respectively. Thus, one mahāyuga lasts 4,320,000 human years (Danielou's figure of 2,160,000 [p. 249] is in error.).

⁵⁰ One cycle of yugas = 1 mahāyuga; 1000 mahāyugas = 1 kalpa or 1 day of Brahmā. Brahmā lives for 100 years; at the final moment of the 100th kalpa, the cosmos—as well as all the gods and sages and Brahmā himself—cease to exist (see Danielou, p. 249). Interestingly, the kalpa which is defined mythologically as one day of Brahmāa is also defined by ancient Hindu astronomers as the interval of time between two consecutive instances of the common conjunction at the zero point of the Hindu zodiac (the first point of Āśvina) of all known heavenly bodies and astrological influences, i.e., of all nine Hindu planets as well as their apogees or aphelia and their orbital nodes (see So-mayaji [n. 1 above], ch. Ill, No. 1).

⁵¹ Patterns in Comparative Religion (New York: Meridian Books, 1966), p. 154.Google Scholar Also see his “Cosmical Homology and Yoga,” Journal of the Indian Society of Oriental Art, V (1937), pp. 197–200.Google Scholar

⁵² See Eliade, Patterns (n. 51 above), pp. 159, 161–63.

⁵³ See Pillai, G. K., Hindu Gods and Hidden Mysteries (Allahabad: Kitab Mahal, 1958), p. 53; confirmed in interviews.Google Scholar

⁵⁴ For a treatment of powers of beings as a function of the exchange of coded substances, see the work of Marriott, M. and Inden, R. (“Caste Systems,” Encyclopedia Britannica, 1973, pp. 983ff.)Google Scholar; 1 am deeply indebted to their views on this subject.

⁵⁵ As the chief medium for the transfer of power-giving substances and means of dissipating polluting substances, water and bathing are important in nearly every Hindu ritual, though their importance is heightened in general observance at special times as on amāvasyās. In the specific context of the Khaṇḍobā cult, though water and bathing play an important role at other times, at no time are they as important as on Somavati Amāvasyā, which is the only time in the cult when there is a regularly established procession to the river for bathing of images.

⁵⁶ Conceived of, in the mythology, as nectar.

⁵⁷ Especially by rewarding their vows (navas)— usually involving virility, fertility, or both.