In estimating the value of this piece of evidence it is necessary to remark, in the first place, that the authority is no longer that of Aristoxenus, but of a certain Lysis, of whom nothing else seems to be known. That he was later than Aristoxenus is made probable by his way of describing the Mixo-lydian octave, viz. by reference to the notes in the Perfect System by which it is exemplified (Hypatê Hypatôn to Paramesê). In Aristoxenus, as we shall see (p. 31), the primitive octave (from Hypatê to Nêtê) is the only scale the notes of which are mentioned by name. But even if the notice is comparatively early, it is worth observing that the Mixo-lydian scale thus ascribed to Lamprocles consists of two tetrachords of the normal type, viz. with the semitone or [Greek: pyknon] at the lower end of the scale (Diatonic _e f g a_, Enharmonic _e e* f a_). The difference is that they are conjunct, whereas in the primitive standard octave (_e - e_) the tetrachords are disjunct (_e-a b-e_). This, however, is a variety which is provided for by the tetrachord Synêmmenôn in the Perfect System, and which may have been allowed in the less complete scales of earlier times. In any case the existence of a scale of this particular form does not prove that the octaves of other species were recognised in the same way. (2) In another passage (c. 6) Plutarch says of the ancient music of the cithara that it was characterised by perfect simplicity. It was not allowed, he tells us, to change the mode ([Greek: metapherein tas harmonias]) or the rhythm: for in the primitive lyrical compositions called 'Nomes' ([Greek: nomoi]) they preserved in each its proper pitch ([Greek: tên oikeian tasin]). Here the word [Greek: tasis] indicates that by [Greek: harmoniai] Plutarch (or the older author from whom he was quoting) meant particular _keys_. This is fully confirmed by the use of [Greek: tonos] in a passage a little further on (c. 8), where Plutarch gives an account of an innovation in this matter made by Sacadas of Argos (fl. 590 B.C.). 'There being three keys ([Greek: tonoi]) in the time of Polymnastus and Sacadas, viz. the Dorian, Phrygian and Lydian, it is said that Sacadas composed a strophe in each of these keys, and taught the chorus to sing them, the first in the Dorian, the second in the Phrygian, and the third in the Lydian key: and this composition was called the "three-part Nome" ([Greek: nomos trimerês]) on account of the change of key.' In Westphal's _Harmonik und Melopöie_ (ed. 1863, p. 76, cp. p. 62) he explains this notice of the ancient modes ([Greek: harmoniai], _Tonarten_), observing that the word [Greek: tonos] is there used improperly for what the technical writers call [Greek: eidos tou dia pasôn].