There still remain the _fabulae Atellanae_, so called from Atella, an Oscan town of Campania, and often mentioned as _Osci Ludi_. These were more honourable than the other kinds, inasmuch as they were performed by the young nobles, wearing masks, and giving the reins to their power of improvisation. Teuffel (L. L. § 9) considers the subjects to have been "comic descriptions of life in small towns, in which the chief personages gradually assumed a fixed character." In the period of which we are now treating, _i.e._ before the time of a written literature, they were exclusively in the hands of free-born citizens, and, to use Livy's expression, were not allowed to be polluted by professional actors. But this hindered their progress, and it was not until several centuries after their introduction, viz., in the time of Sulla, that they received literary treatment. They adopted the dialect of the common people, and were more or less popular in their character. More details will be given when we examine them in their completer form. All such parts of these early scenic entertainments as were not mere conversation or ribaldry, were probably composed in the Saturnian metre. This ancient rhythm, the only one indigenous to Italy, presents some points worthy of discussion. The original application of the name is not agreed upon. Thompson says, "The term Saturnius seems to have possessed two distinct applications. In both of these, however, it simply meant 'as old as the days of Saturn,' and, like the Greek _Ogugios_, was a kind of proverbial expression for something antiquated. Hence (1) the rude rhythmical effusions, which contained the early Roman story, might be called Saturnian, not with reference to their metrical law, but to their _antiquity_; and (2) the term _Saturnius_ was also applied to a definite measure on the principles of Greek prosody, though rudely and loosely moulded--the measure employed by Naevius, which soon became _antiquated_, when Ennius introduced the hexameter--and which is the _metrum Saturnium_ recognised by the grammarians." [21] Whether this measure was of Italian origin, as Niebuhr and Macaulay think, or was introduced from Greece at an early period, it never attained to anything like Greek strictness of metrical rules. To scan a line of Livius or Naevius, in the strict sense of the word, is by no means an easy task, since there was not the same constancy of usage with regard to quantity as prevailed after Ennius, and the relative prominence of syllables was determined by accent, either natural or metrical. By natural accent is meant the higher or lower pitch of the voice, which rests on a particular syllable of each word _e.g. Lúcius_; by metrical accent the _ictus_ or beat of the verse, which in the Greek rhythms implies a long _quantity_, but in the Saturnian measure has nothing to do with quantity. The principle underlying the structure of the measure is as follows. It is a succession of trochaic beats, six in all,