. 22 Canadian Arctic Expedition, 1913-18 



that the writer decided to leave this as a special example of rivalling accents 

 and to publish three different transcriptions, any of which sounds perfectly 

 right with little or no mental adaptation. 



All that we have to work with in the present instance is the songs as sung 

 with the words, and we have been obliged to accept the accents as they have 

 fallen under these conditions and to note the actual lapse of time in half beats 

 as well as whole ones, in rests as well as in tones. The extra half beats are quite 

 as often made in actual separate eighth notes as in quarters which are held a 

 little longer than might seem necessary, so they are not only the result of dram- 

 atic singing or want of breath. 



As to tempo, the following lists will prove that there is practically no differ- 

 ence between what the people designate as a pisik and an aton. 



The metronome tempo per quarter-note for the two types is as follows : — 



Aton 



116 



84 



80 



88 



112 



92 



132 



92 



88 



76 



88 



88 



100 



108 



80 



96 



100 



96 



For the pisik without prelude the average is 99, the lowest being 80 and the 

 highest 130. The average speed for the pisik with prelude is 97, the lowest 

 speed 76, and the highest 126. For the aton the average is 95, the lowest being 

 76 and the highest 132. These are all practically the same, so that it would 

 take more than human ability to differentiate the songs on the basis of tempo, 

 especially by people who had nothing more to go by than a mere hearing of 

 them. 



Having thus disposed of the usual points that are of importance with us in 

 the differentiating of dance melodies, we may consider the question of smaller^ 

 rhythms, that is, regular successions of actual note values, but here again, 

 search seems useless. As far as the succession of words is repeated, the rhythms 



