438 SIHLER. 



tive or systematic treatise on physical phenomena both normal 

 and abnormal ; at the first reading of the greater part of Book 

 V and all of Book VI one cannot suppress a feeling that sys- 

 tem is cast to the winds and to miss that rigid, comparativel}' 

 speaking, that rigid sequence of treatment which is so unmis- 

 takable in the general unfolding of Epicurean doctrine in Books 

 I— IV. Cf Munro's commentary on Lucretius V, 533. And 

 with Epicurus' incessant railing against the postulate of one ex- 

 planation (rov yLOvayji zi>67iOv, §95 1. c. and § 1 13 ro (5s fitav 

 ahiav TO'jzioii^j-odcdovac^ TiXiovaycoz tCov (fatvoiievcovvAYM'koup.kviov, 

 fiavrAov i(TTi). Cf. Lucretius V, 620, *' no// inquam, simplex his 

 rebus reddita causast." 



^' 729 [of two different astronomical theories] 

 ^'proinde quasi id fieri nequeat quod pugnat uterque 

 '^aut minus hoc illo sit cur amplectier ausis. " 



And 751 



Solis item quoque defectus lunaeque latebras 

 pluribus e c ausis fieri \S\A posse putandumst. 



And so again in book VI, 703 sqq., the theory of alTtoKojia is 

 advanced even more clearly : 



" Sunt aliquot quoque res quarum nnam dicere causam non 

 satis est, veruni plnris, nneie nna tanien sit ; as f e. when you 

 see the dead body of a man lying at a eiistanee [/. e., preclud- 

 ing a close and direct inspection on our part] ; there it behooves 

 us to exhaust the entire range of contingencies through which a 

 man may perish ; although we cannot, at that distance, prove 

 any particular single one : the sword, or frost, or disease, or 

 poison. And so we find the same plurality of explanation in 

 Lucretius : positive and exclusive asseverations in this sphere 

 are impossible. 



V 526 nam quid in hoc mundo sit eorum ponere certrun diffieile 

 est ; sed quid possit fiatque per omne [das All]. 



in variis mundis varia ratione creatis 



id doceo ////mque sequor disponere causas, etc. . . 



