156 BUREAU OF AMERICAN ETHNOLOGY [BuLy, 75 
78.) Six renditions were recorded in 1914, the renditions being — 
uniform and connected without a break in the time. The transcrip- 
tion is from the cylinder made in 1914. Five renditions of the same 
song were recorded in 1916, at which time the writer did not play 
the first record, but ‘“‘hummed” the transcription. The singer recog- 
nized it and said he would like to make another record of it. This 
second cylinder was transcribed and the results compared. In this 
comparison it is found that all the renditions recorded in 1916 omit 
the fifth and sixth measures and the last measure of the song as 
transcribed. As these measures are repetitions, this would be of 
slight importance, but as it reduces the occurrences of these phrases — 
from three to two the change suggests a tendency toward regularity 
——= 
oie Wag am S 
* * 
Ve x x 
\ 
+ Pan = eos 
x x 
x 
+ x 
x 
x x x x 
gle 
¥ x 
# * 
jibe 
a 
Fiq. 13.—Diagram, Scalp dance. a, Singers with hand drums. 6, Two circles of women moving in 
opposite directions. 
and conventionality. The first group of renditions was in a tempo 
of J=104, and the key of G minor; the second was faster (j= 108) 
but in a. lower key—the key of F sharp minor. The 5-4 division of 
certain measures was strictly maintained. The song contains the 
tones of the second five-toned scale and is melodic in structure. 
Considering the lack of variety in progressions the melody is inter- 
esting and quite diversified. One-half the intervals are minor thirds, 
and of the other half all except one interval are major seconds, that 
interval being an ascending fourth. Other songs containing rests are 
noted in the analysis of No. 8. 
The dancing with the scalps (commonly known as the Scalp dance) 
began late in the afternoon and ended soon after dark. There was 
no feast connected with this dance. The singers stood im a line facing 
