[1385]. See e. g. Crooke, Folklore of Northern India, ch. 5 (the Black Art), and especially pp. 264 foll.
[1386]. See e. g. Leland, Etruscan Roman remains in popular legend, pp. 3 and 195 foll.
[1387]. The chief attempts are those of Unger, in Rhein. Mus., 1881, p. 50, and Mannhardt in his Mythologische Forschungen, pp. 72-155. The former is ingenious, but unsatisfactory in many ways; the latter conscientious, and valuable as a study in folk-lore, whether its immediate conclusions be right or wrong. See also Schwegler, R. G. i. 356 foll.; Preller, i. 387 foll.; and article s. v. in Dict. of Antiquities (2nd edition); Marq. 442 foll. The ancient authorities are Dion. Hal. 1. 32. 5, 79, 80; Ovid, Fasti, 2. 267 foll.; Plutarch, Caes. 61, Rom. 21; Val. Max. 2. 2. 9; Propert. 5. (4.) 1. 26; and many other passages which will be referred to when necessary.
[1388]. Dion. Hal. 1. 32. 5.
[1389]. Jordan, Kritische Beiträge, 164 foll. Unger’s attempt, after Serv. Aen. 8. 343. to derive the word from luo (‘to purify’) is generally rejected.
[1390]. Wissowa, Lex. (s. v. Lupercus) takes the latter view, but rightly, as I think, rejects the deity.
[1391]. Virg. Aen. 8. 630 ‘Mavortis in antro.’ Roscher, in Lex. s. v. Mars, 2388; Preller, i. 334.
[1392]. Plut. Rom. 21. After mentioning the goats, he says, ἴδιον δὲ τῆς ἑορτῆς τὸ καὶ κύνα θύειν τοὺς Δουπέρκους (cp. Q. R. iii).
[1393]. Marq. 165. See above, p. [110].
[1394]. So Val. Max. l.c. From Ovid’s version of the aetiological story of Romulus and Remus (Fasti, 2. 371 foll.) we might infer that the feasting took place after the running.