When we read the Latin Grammarians' Rules of Prosody, we are puzzled now and then. (They are to be found in the Prosody-sections of Keil's Grammatici Latini, and deal mainly with the Dactylic Hexameter and with Virgil.) One thing that puzzles us is their silence about the features of difference between Latin Prosody and Greek. They often seem to take it for granted that Virgil's Prosody is identical with Homer's. This point of view is perhaps not surprising, since these Grammatici often speak of Latin as a mere dialect of Greek (Charisius 292, 16 K. ‘cum ab omni sermone Graeco Latina lingua pendere uideatur’; Diomede 3II, 3 ‘cum ab omni sermone Graeco Latina loquella pendere uideatur’). But it has its disadvantages. Every scholboy knows that moeniă Troiae is as natural in Virgil as TєίεΧεă Τροίηs would be unnatural in Homer; and every school-manual of Latin Prosody confines its examples of a Mute and Liquid (1) lengthening, (2) not lengthening a preceding syllable to examples of a Mute and Liquid in the middle of a word. If it mentions Catullus' impotenti̅ freta, it calls this a Greek, not a Latin type. Not so the Grammatici. Diomede's examples of a short syllable before FR, FL are (429, 3 K):

ore fremebant talia flammato.