Remarks on some of the Greek Tragic Fragments. 227 TvpavvlSos r ayovcriv alcrxLcrTTju edpav. eTretra 5' ovdels %6pos ovre (pverat 7rpos xPW a ^ i T f (pvvTes apvovvrai arrvyeip. $eiv6s yap epTreiv ttXovtos es re ra/3aTa Kal Trpbs ra. /3aTa, x a)ir @ V ^ (vr ] s aVT )P /Lir/S' ivTvxoov bvvair' av <v ipa rvx^v. Kal yap 8vo~ei8es o~a>p.a nal dvacovvpov yXcoccrrj o~o<pbv TiBrjcriv evpopcpou t' iSeii/. p.6v(o de x a ^P tv Kai vocrelv iov<rla Trapeariv avrS Kairucpv^racrOai kclku. In v. 3 there is considerable variety of reading in the MSS. of Stobaeus, i^Kovcriv or aKOvaiv being found for t' ayovatv, dyxloTrjv and f)diaTT)p for alo-xLo-Trjv. dyxia-rrjv would seem to be right, as furnish- ing the best explanation of alo-xio-rrju, out of which no alteration in the context could extract any sense. If we take ayx^ Tr l v with rvpavvLSos, Salmasius' OaKova-iv for r ayova-iv is probable enough, though something might be said for avoiding a change of con- struction, by reading rayeia-av. Possibly, however, the word may be a corruption of a substantive in the dative, which was meant to be constructed with dyxtcmjv, a structure found in Homer and Pindar ; and in that case the word can hardly have been anything else than deolo-iv. In v. 7 /3ara appears to be merely an error for fido-ipa. The strange use of bvo-awpov, v. 9, as opposed to <ro<p6v, probably requires noting and illustrating, not correcting : but the word may be a corruption of some less known compound, such as dvo-TcopvXov, or dva- may have crept in from 8vo-eibes, the text having been originally something like ko yap dva-eibes aSpa ko. TTjrcopevov ya>crar)s o~o<fi6v ridrja-iv. In V. 11, Bergk's Kavocrelv and Ellendt's kov voa-fiv rather jar with KdniKptyao-dai, so that we must either accept Vater's explanation x a ^p lv *a! voae7v = x a lpeiv voo-ovvto, or read voa-ovvr f where the mixture of datives and accusatives would be idiomatic enough. With the sense comp. Juv. vn. 193, " Felix orator quoque maximus et jaculator : Et si perfrixit, cantat bene." Soph. Amphiaraus. fr. 8 (124). %v& ovre rreWeis oi* aypavXos {Soros* It seems evident from the passage in Erot. Lex. p. 306, that Schneidewin and others are right in supposing ttsKKcis to be a mis- take for some case of TrcXXdy, probably iriKKrjs. It is strange how- ever that Schneidewin, who himself refers to rap oh rhv ncKKdv, Theocr. v. 99, should have changed 6V here into pw6t>, when of so 152
