178 PROCEEDINGS OF THE AMERICAN ACADEMY. 



the Persians were not going to attack until the signal was actually 

 given ; for how could they know when it was ready to be given ? If npiv 

 had been used the form would probably have been nplv av <pavy. 



One of the clearly marked and distinctive characteristics of Ilerod- 

 otus's style is his frequent use of the infinitive in subordinate clauses in 

 oratio obliqua. The general fact has long been known, but an exami- 

 nation of the temporal clauses in Herodotus shows that this usage is 

 surprisingly frequent. The infinitive is found in a temporal clause 46 

 times, as follows * : eirel — 4, 10, 33; 5, 84; 7, -150. eVeiS^ — 3, 26. 

 fain — 1, 94; 2, 42, 118; 6, 84, 137; 8, 135. isS—l, 94, 202; 

 2, 102; 5, 85, 86. eVre — 7, 171. *~a>9 — [4, 42]. t 6'kok — 2, 140. 

 e V™ — 3, 105. rrpiv — 1, 165. is— 1, 24, 86, 94. 141 ; 2, 99, 107, 

 121/3, 121/3,, 121/3 ;s , 1218, 1218,, 121e, 140, 162; 3, 35, 87; 4,8, 9, 11, 

 179; 6, 137; 7, 148, 170; 8, 94, 94.,, US. 



In order to find out the proportion between the cases where the infini- 

 tive was used and those where the finite mood was retained, I collected 

 the passages where a finite verb in a temporal clause in oratio obliqua 

 might, according to the principles in the cases above enumerated, have 

 been changed to the infinitive but was not so changed. There are 65 

 such cases. Out of 111 cases, then, where it was possible for Herodotus 

 to make this change, in 46 cases he used the infinitive, in 65 he did not. 

 The infinitive so used stands for an indicative 43 times, for an optative 

 once (2, 140), and for a subjunctive twice (1, 165, 202). 



In 2, 140 : okcos yap oi (poirdv . . . Alyvirriav its {Kclittokti TrpoaTerd^dai 

 . . ., is rr)v 8u>pei]v KtXeveiv crcpeas kci\ an»86v Koptfctv, (poirav clearly stands 

 for (poiTo>(v of the oratio recta. In 1, 202: ifKe vvos 8i iTrifiaWoptvov tov 



Kapnov paWov peOvaKeadai, i s o is opxTjo-lv re dviaTaadai Kal is 0018171' cnnKvt€<j6ai, 



the main verb expresses generic action, and we should expect after is o 

 the abrist subjunctive with av. t In 4, 43 (e's 6 av dnUTiTai) and 8, 108 

 (is o e\6>]),(s o with the subjunctive stands in oratio obliqua; but the sub- 

 junctive is not, as here, changed to the infinitive. One reason for the 

 change in 1, 202, is this: is has here dropped its distinct temporal 

 force and is virtually equivalent to a mere co-ordinating conjunction. 

 Cf. 4, 203 (is 6 . . . peTeptXrjo-f), where no main verb is expressed. The 

 meaning of our passage is this : " as more of the fruit is thrown on they 

 drink more and more, and finally they begin to dance and sing." 



* The list of Ileilmann, De Infinitivi Syntaxi Herodotea, Gissae, 1879, pp. 49 f, 

 is not entirely accurate. 



t See below, pp. 179 f. J Goodwin, Moods and Tenses, 613, 5. 



