60 BUREAU OF AMERICAN ETHNOLOGY [bdll. ei 



received them in dreams, or who have inherited or purchased the 

 right to sing them. Such songs form one class of Indian music. 

 To another class belong songs of games and dances, songs of war 

 and of the hunt, as well as songs which celebrate the deeds of a 

 chief or successful warrior. The words of all these songs, while often 

 showing high degree of poetic feeling, appear to be less subjective 

 in character tlian the words of similar English songs. It is difficult 

 to state what Indian songs correspond to the folk songs of the white 

 race, but in such a group should probably be included the songs 

 connected with folk tales. These are many in number and usually 

 are represented as being sung by animals. An example of this is 

 the story and song of the crawfish (Bulletin 53, p. 305). Songs of 

 this class are found to vary more than any others in their rendition 

 by different people ; perhaps because they were more generally known 

 and sung than others, with less criticism as to correctness of rendition. 



Among musicians of civilized races the standards of excellence in 

 a singer include (1) intonation, (2) quality of tone, (3) range of voice, 

 (4) memory, and (5) interpretation (intellectual and emotional). 



(1) Intonation. — Qbserving Indian singers according to these 

 standards, the writer has noticed that "good singers" keep the pitch 

 of their tones approximately that of the tones of the diatonic scale, 

 and that in songs recorded by such men and women the pitch of 

 " accidentals " is practically the same in every rendition of the song. 

 This is not the case in songs phonographically recorded by men whose 

 musical standing among their own people is not so high. An instance 

 of this is as f oUows : On the transcription of a song recorded by an old 

 man it was found that the melody showed no feeling for a keynote. 

 Inquiry among the Indians disclosed the fact that the man was not 

 considered by them to be a good singer. The same song was accord- 

 ingly obtained from a man of acknowledged musical proficiency, and 

 on comparing the renditions it was found that by slight alterations 

 in pitch the song had become diatonic in character, the intervals 

 closely resembling those of the scale and ending on a keynote. Ex- 

 perience in listening to the melodies of Chippewa and Sioux makes it 

 possible to distinguish between a melody which has unusual native 

 peculiarities and one which is distorted by a poor singer. 



The accuracy of an Indian in repeating a song should also be con- 

 sidered. For this purpose several consecutive repetitions of each 

 song were recorded. As an evidence of accuracy in repeating cere- 

 monial songs the following incident may be cited: In 1912 the writer 

 recorded four songs of the Creek Women's society of the Mandan, 

 from ;Mrs. Holding Eagle, one of its members. In 1915 ^Irs. Holding 

 Eagle recorded the songs a second time, and on comparison it was 

 found that the pitch and metronome speed of all tiie songs was the 



