δέδια. If in the whole range of the Greek language some half a dozen instances of a distinct perfect middle can be found, in addition to the perfect active, surely this is no adequate ground for representing these two tenses as the proper and regular complement of the verb, unless it be proper to confound the rule with the exception.
We now come to the consideration of the passive tenses. It is undoubtedly much easier to produce duplicates here, than either in the active form or the middle. Examples such as ἀλλαχθῆναι and ἀλλαγῆναι, συλλεχθῆναι and συλλεγῆναι are by no means scarce, even among the Attic prose-writers. But in the first place the difference of formation is in these cases slight, not at all affecting the terminations; and secondly there is not a shadow of ground for supposing that these two forms were used as distinct tenses, that is, with any difference of signification: on the contrary it is evident that they were perfectly equivalent, and used, as regarded their signification, with absolute indifference; in short they were mere varieties of formation, which, in verbs of a certain description, got into use for one and the same tense. This is evident from the circumstance, that hardly any author will be found using more than one of these forms in the same verb: the fashion, so to speak, by which a preference was given to one or the other, having prevailed at different times and places. Moreover the proportion of verbs in which even this, the most numerous, species of duplicates obtain, is very limited, being confined almost entirely to a portion of those in which labial or guttural consonants precede the final ω. It is only therefore, at most, to verbs of this particular class that any rule for their formation should extend. But the fact is, that, although writers of different ages and dialects formed these tenses with some variety, yet any given writer seldom felt himself at liberty to use more than a single form. To revert once more to our own language, the case is simply the same as with I spake and I spoke, I brake and I broke, I catched and I caught, where an older and a newer form occur in writers of different ages or styles, but still most strictly as representatives of the same tense. To found on such anomalies a superfluous complexity in the general mechanism of a language, and especially to introduce such complexity into
