BOOK I CH. XIx § 49. 143

tenere diceret. Lucretius speaks in equally high terms of his master’s speculations, 1 74 omne immensum peragravit mente animoque foll.
docet eam esse vim—aeterna. This extremely difficult passage has been discussed by many writers, esp. by Sch. Opuse. 111 3151, and Neue Jahrb. for 1875 pp. 687—691, as well as in the notes and app. to his ed. ; but the first to give a satisfactory explanation of the whole was Hirzel in his Untersuchungen pp. 46—90. He translates as follows (p. 68): ‘Epikur lehrt die Natur der Gétter sei der Art, dass sie erstens nicht mit den Sinnen, sondern nur mit dem Geiste erfasst wird, und dass sie ausserdem weder Soliditiét noch individuelle Identitit besitzt, wie die sogenannten orepewma ; vielmehr gelangten wir zur Erkenntniss des Gittlichen (denn das besagen die Worte quae sit et beata natura et acterna) durch Bilder, die wir wahrnehmen’ &c.- I had long taken the same view of the construc- tion of capere, and of the needlessness of Sch.’s emendations cernantur, cumque, beatae naturae. The clue to the right interpretation is to be found (1) in § 105, where the account here given is criticized by Cotta, and (2) in Diog. L. x 139 é€v Gddois S€ hyot rods Geots Ady@ Oewpnrors, ods pev Kar’ apiOpov vpectaras, ods S€ ka Cuoediav ex THs Tuvexovs emippicews TOY Opolwy eidadov emt Td atTO amoreteNeopevayv avOpwroedas. Philodemus seems to have treated of the subject in his wepi evoeSeias, but unfortu- nately the passages relating to it are too corrupt to afford much help. See p. 110 duvarar yap ek Tis dpordtntos (similitudine) vmapyovea (ididrns) Stardviov fxew Thy Tedelav evdampoviav, emevdnmep ovX WTTOv ek TAY avTav i} TOV Opoiwy oTotxeiwy Evornres arroreheioOat Svvavrac. Gomperz despairs of the passage (see his n. ‘ dieser mir zum kleinsten Theil verstindlichen Columne’,) but it would appear to be a comparison between our ordinary modes of perception and the mode in which we arrive at a consciousness of deity ; just as in the next page it is said ‘if opponents charge Epicurus with denying the existence of the Gods, why might they not on the same ground charge him with denying the existence of horses and men, kal rav@? a7@s Ta kata pépos alcOnra Te kai vonra dicey €idn?’? The same subject is discussed in pp. 132—138, but only occasional phrases are legible, as ryy kat’ dpiOuov avyxpiow (C.’s ad numerwm) in pp. 134 and 138, prre yap dropous vouitew Tors Oeovs pte ovvOérovs p. 136, apparently an exhaustive argument to prove the atheism of Ep. ‘his Gods are neither atoms nor compounds of atoms, and what other entities are admitted by him ?”
non sensu sed mente cernatur: cf. Lucretius quoted on quasi corpus, and § 105 speciem dei percipi cogitatione non sensu. Sch. points out that while L. speaks of the tenuis natura and Cotta of the species, both referring only to the fine etherial body of the Gods, Vell. speaks more generally of vis et natura. This is because he is about to refer, not merely to the
1 He calls it locum omnium dificillimum cujus certam omnibusque numeris absolutam interpretationem vix quisquam, ego certe hoc tempore proponere non possum.
144 BOOK I CH. XIX § 49.
immediate sensuous impression produced on the mind when its fine atoms are struck by the cognate atoms which constitute the divine imagines, atoms which pass unperceived through the coarser sieve of the bodily senses, but also to the conception of blessedness and immortality to which the mind attains by reflecting upon the impressions it has received. It is the latter process which is properly expressed by cogitatio.
nec soliditate—appellat. At first sight it seems natural to take sol. as an abl. of cause after cernatur ; and so Sch. explains it by a reference to the distinction between the zmagines thrown off from solid bodies (the ore- pépua), Which tmagines are described in Diog. L. x 46 as droppora thy €&qs Oéow kat tak Scatnpotoa, jvrep Kat ev Tois oTepepviots eixov, and a finer class of imagines which reveal to us the shadowy form of the Gods. The expression would not be quite accurate, for even the finest images must in the end consist of atoms (since all that exists is summed up under atoms and void, according to Epic.) and soliditas is essential to atoms of every kind ; still in popular language (quadam=ut ita dicam) it might be said that the images perceived by the bodily senses were perceived in virtue of a massiveness which was not shared by the images which were per- ceptible by the mind alone. The objections to this interpretation are (1) that it really adds nothing to what has been already said in the previous clause, though apparently contrasted with it by the word primum, (2) that it is difficult to connect it with what follows, (3) that it is incon- sistent with the words of § 105 xec esse in ea (specie) ullam soliditatem, neque eandem ad numerum permanere, in which the absence of soliditas is predicated of the divine form itself, not of the image, as distinguished from the form, in virtue of which negative property the image is perceived in a particular way. Accordingly Peter (Commentatio de NV. D. Saarbriicken 1861) and Hirzel take soliditate quadam as a predicative Abl. of quality, of which the former cites several exx. (V. D. 112 veris falsa adjuncta tanta similitu- dine, 28 continente ardore lucis orbem, 81 reliquos deos ea facie novimus, 84 his vocabulis esse deos facimus, 107 imagines ea forma, Liv. xxi 62 in agro Amiterno multis locis hominum specie procul candida veste visos nec cum ullo congressos, where there is the same accumulation of ablatives as here) and further illustrates by the following parallel 72 somnis mihi oblata est imago leonis, ut non oculis sed mente cernerctur, neque tngenti corporis mag- nitudine neque densa juba, sed ferocitate oculorum splendore prodita, Taking soliditate thus as referring to the substance of the deity which has nihil concret?, nihil solidi in it (§ 75), it is opposed to the previous clause which referred to the mode in which that substance was perceived. It cannot be denied that there is something very harsh in the construction of such an Abl. with cernatur, and I think it possible that s’¢ may have been lost after numerum before wt. The term orepéunoyv occurs repeatedly in the frag- ments of Epic. wept duces and in his Epistle to Herodotus preserved in Diog. L. x.
We come now to the more difficult ad numerum, which must evidently