This can hardly be done until we make up our minds that the subjects of education must not be increased in number, and that moreover they may be alternated with far more freedom than is now the case. There is, for example, a superstition that everybody must learn Latin before learning Greek, and that French is a sort of necessary accomplishment for a lady, whereas it is perfectly certain that the cultivation to be attained through Greek is ten times as great as that we can gain through Latin; while in the second case it is no paradox to assert that any woman able to understand the _Antigone_ of Sophocles or the _Thalusia_ of Theocritus would derive from them more spiritual food than from all the volumes of French poetry she is ever likely to read. If we cannot compass all, the lesser should give way to the greater; and it is not till our own day that the supremacy of Greek has been acknowledged by all competent judges. [Sidenote: The nature and quality of Roman imitations.] § 97. What has promoted the reign of Latin, and has told against Greek in our schools, is partly, I believe, the bugbear of a strange alphabet; partly also--and this among more advanced people--the want of a clear knowledge how closely most Roman poetry was copied from Greek models. Were the Greek models now extant, the contrast would probably cause the Roman imitations to disappear, as indeed many such must have disappeared when the Roman readers themselves approached the great originals. Even now, if the lyrics of Sappho and Alcæus were recovered from some Egyptian tomb or from the charred rolls of Herculaneum, it might have a disastrous effect on the popularity of Horace. [Sidenote: The case of Virgil.] But in most cases the Romans copied from inferior poets of the Alexandrian age; and before the reader and I part company, it is of importance to insist upon this,--that the best of Roman poetry was often a mere version of third-rate Greek. By far the greatest of the Roman poets is Virgil; and if he alone remained, Latin would be worth learning for his sake. But even Virgil copies from second-rate Alexandrian poets, Apollonius and Aratus--from the latter to an extent which would be thought shameful in any independent literature. It may be true that the translations are in this case not only equal, but far superior, to the originals. I will not dispute this, as my case does not require any doubtful supports. For even granting that he can exceed a second-rate Greek model, what shall we say when he attempts to imitate Theocritus in his _Bucolics_? Here he is taking a really good Greek poet for his model, and how poor is the great Roman in comparison! Even therefore in imitating an Alexandrian master, we can see that the first of Latin poets cannot bear the comparison. [Sidenote: Theocritus only a late flower in the Greek garden of poetry.]