18 BUREAU OF AMERICAN ETHNOLOGY [bull. 45 



minor third unci ji;lissand() progressions, and lias a compass of an 

 octave and two tones. 



Many of the MIde' songs are grouped in series, the pictures being 

 drawn on a long strip of birch bark. Midway through such a series 

 a hn(> is drawn. This indicates a pause, after which the people rise, 

 and dance during the remaining songs of tlie series. 



'I'lie anticjuity of these songs is shown l)y the fact that many of 

 tliein are widely known among scattered peoples who came originally 

 from the same locality but have had practically no communication 

 for a long period of time. The })reservation of the songs is aided by 

 the mnemonics, and the importance attached to them is shown by the 

 reluctance of the Mide' Indians to sing them until fully assured of the 

 sincerity and good will of the person making the rec{uest. 



In analyzing the music as the important part of the expression, 

 the (juestion arises whether the melodic progression or the rhythm 

 is more strongl ,' impressed on the mind of the singer. Which of 

 these elements is to him the more important feature of the song? 

 A careful review of more than a hundred Mtde' songs shows them to 

 consist of simple intervals and complicated rhythms. The tones 

 comprised in the songs are liniit(><l in nund)er, many of the songs con- 

 taining onl , three or four tones, except as the number is extended 

 by repetition in a lower octave; the variety of rhythms is great, as 

 will be seen in the transcriptions. Accidentals rarely occur in the 

 songs. An accidental in the o])cning measures of a song is worthy 

 of little consideration, as in many instances tlie introductory measures 

 are sung oid ,• once, and the singer is allowed some freedom in them. 



The rhvthm of the song is determined by noting the accented 

 tones and dividing the song into measures according to them. 



By observation we fmd that in many of the songs the metric unit 

 is the measure, not the individual count in the measure. In these 

 instances the accented measm-e beginnings are found to conform to 

 a very slow metronome beat, but the intervening tones are irregular 

 in length and can not be accurately indicated by note values. These 

 songs would resemble chants except for the freedom of their melody 

 progressions. 



The next fact which we note in connection with the rhythm is that 

 it is most peculiar in songs which are intended to produce magic and 

 mysterious results. In this class are included songs for the healing 

 of the sick as well assimgs used in ccmnection with special "medicine" 

 for success in hunting or other mulertakings. In many of these songs 

 there is no repetition of a rhythmic unit, but the entire song consti- 

 . tutes a rhythmic nnit. its repeated renditions forming a satisfactory 

 whole. 



There are other songs in which two or more measures of varying 

 lengths combine to form a rhythmic unit , which is repeated throughout 



