"De Musicâ" at about the time that St. Ambrose wrote his hymns, gives us an idea as to the form which the melodies must have had originally. He defines music as "the science of moving well" (scientia bene movendi) and the Iambic foot as consisting "of a short and a long, of three beats". As in the case of St. Ambrose we have poet and composer in one person, it is but natural to suppose that his melodies took the form and rhythm of his verses. The fact that these hymns were intended to be sung by the whole congregation, over which, according to the Arians, the saint cast a magic spell by means of his music, also speaks in favour of their having been syllabic in character and simple in rhythm. For several centuries it has been held that St. Ambrose composed what are now termed antiphons and responsories. There is no satisfactory proof that such is the case. The fact that he introduced the antiphonal (alternate) mode of singing the psalms and his own hymns (each of the latter had eight stanzas), by dividing the congregation into two choirs, probably gave rise to this opinion. The responsory as practised by direction of St. Ambrose consisted in intoning the verse of a psalm by one or more chanters and the repetition of the same by the congregation.
Guido Maria Dreves, S.J., F. A. Gevaert, Hugo Riemann, and others have endeavoured to show how the melodies belonging to the authentic Ambrosian texts have been transmitted to posterity and what rhythmical and melodic changes they have suffered in the course of time in different countries. Dreves first consulted the "Psalterium, cantica et hymni aliaque divinis officiis ritu Ambrosiano psallendis communia modulationibus opportunis notata Frederici (Borromeo) Cardinalis Archiepiscopi jussu edita. Mediolani apud hæredes Pacifici Pontii et Joannem Baptistam Piccaleum impressorem archiepiscopalem, MDCXIX" and the complete Ambrosian manuscript Hymnary in the Bibliotheca Trivulziana in Milan, which two works are most likely to contain the best traditions. The melodies as they appeared in these works were then compared with manuscripts of the twelfth, thirteenth, fourteenth, and fifteenth centuries at Naples, Monza, Prague, Heiligen Kreuz, St. Florian (Austria), Nevers (France), and Coldingham (Scotland), preserved by the Cistercian monks, who from the foundation of that order had used the Ambrosian hymnary and not the Roman. This comparison made it possible to eliminate the many melismatic accretions and modifications received, evidently, at the hands of singers who were influenced by the taste of their times and found the original melodic simplicity unsatisfactory. As to the rhythm, it must be remembered that the Ambrosian, like all plain-chant melodies, lost their rhythm in the course of the Middle Ages. They were transcribed from the ancient neumatic notation into square notes of equal length, the time given to them being determined by the text syllables to which they were sung. Bearing in mind St. Augustine's definition, and the fact that in St. Ambrose's time accent had not overshadowed quantity in poetry, we see that Dreves is justified in his mode of restoring the melodies, at least as far as their rhythm is concerned. Inasmuch as all the hymns are written in the same metre, the melodies may be, and undoubtedly have been, used interchangeably. The following illustrations will give us an idea of the different forms of the same melody in the various codices. The melody to the hymn "Æterne rerum Conditor", according to the above-mentioned Psalterium and the hymnary of the Bibliotheca Trivulziana, we reproduce under (a). Under (b) we will give the same tune as it is contained in a codex of St. Florian dating from the fourteenth century. Under (c) is the same melody as restored by Dreves, stripped of its added notes, and in the rhythmical form which it probably had originally.
The hymn "Splendor paternæ gloriæ" exists in more different forms than the one which we have considered above. Version (a) gives the form of the melody as it reads in the Psalterium; (b), as it is in the antiphonary of Nevers of the twelfth century; (c), the version contained in a codex of the thirteenth century in the National Library at Naples; under (d), as it is found in an antiphonary of the fourteenth century in St. Florian, Austria, and, finally, (e) gives us the restored and, probably, the original form.