A century or so before the science of part writing had taken root in musical practice, a strange barbaric form of music meets our eyes. A manuscript of the tenth century, formerly ascribed to Hucbald of St. Armand, who lived, however, a century earlier, gives the first distinct account, with rules for performance, of a divergence from the custom of unison singing, by which the voices of the choir, instead of all singing the same notes, move along together separated by octaves and fourths or octaves and fifths; or else a second voice accompanies the first by a movement sometimes direct, sometimes oblique, and sometimes contrary. The author of this manuscript makes no claim to the invention of this manner of singing, but alludes to it as something already well known. Much speculation has been expended upon the question of the origin and purpose of the first form of this barbarous orgunum or diaphony, as it was called. Some conjecture that it was suggested by the sound of the ancient Keltic stringed instrument crowth or crotta, which was tuned in fifths and had a flat finger-board; others find in it an imitation of the early organ with its several rows of pipes sounding fifths like a modern mixture stop; while others suppose, with some reason, that it was a survival of a fashion practised among the Greeks and Rornans. The importance of the organum in music history has, however, been greatly overrated, for properly speaking it was not harmony or part singing at all, but only another kind of unison. Even the second form of organum was but little nearer the final goal, for the attendant note series was not free enough to be called an organic element in a harmonic structure. As soon, however, as the accompanying part was allowed ever so little unconstrained life of its own, the first steps in genuine part writing were taken, and a new epoch in musical history had begun.