[34] The Lares are thus represented on a coin of the gens Cæsia. See Cohen, _Méd. Cons._, pl. viii, _Cæsia_. [35] Ovid, _F._, v. 129-147. [36] De la Saussaye, _Religionsgeschichte_, i. 281. [37] Wuttke, § 755. [38] Waldron's _Isle of Man_, p. 103. [39] _Encyc. Britan._, art. "Sacrifice." [40] Frazer, _G. B._, i. 239. [41] _Daoine Shie_ or _Sluagh Maith_. [42] Frazer, i. 122. [43] De Rialle, i. 190. [44] See _The Secret Commonwealth_ by Mr. Robert Kirk, Minister of Aberfoyle, 1691. [45] See Saupe's edition of the _Indiculus_, p. 9. [46] Servius on Georg., i. 302, and Prudent, _c. Symm._, ii. 444. [47] Frazer, ii. 332. [48] Preller, _R. M._^3, ii. 198. [49] Cic. _de Div._, i. 18, 36; Plut. _Ti. Gracch._, i. A similar story is related of D. Laelius, _Jul. Op. seq._ 58. [50] _H. N._, xxix. 72. [51] _C. I. L._, i. 603. [52] We have no direct evidence of this, but we may infer it from the analogous case of Dius Fidius:—"Qui per Dium Fidium iurare vult, prodire solet in compluvium."—Non. Marc., p. 494, quoting Varro. The temples of Dius Fidius had a hole specially made in the roof ("perforatum tectum," Varro, _L. L._ v. 66), under which one might swear. Probably the temples of Hercules were similarly provided; certainly those of Terminus were ("exiguum templi tecta foramen habent."—Ov. _F._ ii. 672). [53] Reifferscheid, in the _Annali dell' Instituto_ for 1867, p. 352 _ff._, identifies Hercules with the _genius Jovis_. But, in the first place, this seems to me the wrong inference from his own facts, which all have exclusively to do with the _genii virorum_. Next, the _genius Jovis_ is not known before B.C. 58. Schwegler, before Reifferscheid, noticed that in Gellius, xi. 6. 1, "der römische Hercules erscheint als identisch mit dem genius der Männer."—_R. G._, i. 367 _n._ [54] Roscher's arguments to show that Juno is the moon are not satisfactory. He assumes without proof that Juno was always Lucina (whereas Lucina was an independent spirit worshipped in woods, _Lexikon_, pp. 583 and 602), that Lucina was the moon (whereas she is the spirit that brings children to light, and is not = Luna), that the Italians connected the moon with child-birth (which, as Birt says, lacks proof), that the name _Juno_ indicates a light-giving deity (whereas, though from the root *Div, it does not imply the giving of light any more than _deus_ does, which is applied to the _di manes_, the _di indigetes_, _dea bona_, _dea dia_, &c.). The arguments drawn by Roscher from works of art are untrustworthy, because borrowing is specially probable in their case. Finally, the hypothesis of a Græco-Italian period, on which Roscher relies to prove that Juno = Hera = the moon, is now discredited.