406 PICURIS CHILDREN'S STORIES [eth. ann. 43 



No. 1. In Song No. 2 the tliree phrases ABC together contain five 

 sections, a a in A, b b in B, and c in C, as against seven sections for 

 the ABC phi'ases of No. 1, where a a' x constituted the A phi'ase, b 

 y the B and b z the C. Thus in Song No. 2 the tendency to binary 

 structure in the smaller divisions is rather well marked. Throughout 

 the song the two sections for each A phrase are more often identical 

 than not. This is true also of the two b sections. The binary prin- 

 ciple for the substructure is carried still further in that each section, 

 even c which alone constitutes the C phrase, contains only two meas- 

 ures, but the c compensates for its lack of a complementary section 

 by the extra length of its measures. Nevertheless in actual number 

 of beats the C phrase is shorter than either A or B, although in Song 

 No. 1 the A's, B's and C's differ in their length from time to time, 

 and this is true in the other renditions, Nos. 15, 16 and 17. There is 

 no metric pattern; that is, the alternation of different lengths of 

 measures is irregular and is not consistently repeated, although 4, 

 33^ occurs more often in succession than other combinations. The 

 rhythmic groups conform to the sections. Using numbers for them, 

 the formula is 1, 1, 2, 2, 3, and this is repeated more or less exactly 

 as many times as A B C are given, but the duplication of 1 and 2 

 within the formula is due to repetition of both melody and words 

 and not to rhythmic impulse alone. 



An examination of Nos. 15, 16, and 17 (pp. 429-434), which are the 

 other renditions of No. 2, will show that in larger structure they are 

 more nearly alike than the versions of Song No. 1, but that in most 

 respects the same kind of minor differences from rendition to rendi- 

 tion are to be found. It is rather remarkable, however, that in all 

 three of them the second and fourth A phrases begin on a low note in 

 contradistinction to the relatively high beginning of the first, third 

 and fifth A phrases, confirming the assumption made regarding Song 

 No. 2, that the song has a larger structure than that of phrases, 

 namely, that designated by the roman numerals. 



Song No. 3 (p. 341) appears on first glance not to conform as well to 

 the ternary structural plan as Nos. 1 and 2, for, after two presentations 

 of a set of A B C phrases in the second of which the C is cut short, 

 the song ends with a new musical idea and new words, carried to the 

 length of a phrase or more, to which I have given the letter D. From 

 the standpoint of melody D is nothing more or less than a long exten- 

 sion of what was the third measure in the first full-length C phrase, 

 but its association with new words and its length seem to indicate 

 that it should be considered apart as a distinct phrase rather than as a 

 long-drawn-out continuation of the second C. The text, if not the 

 melody, shows that this D part definitely shortens the second C 

 phrase by cutting into it and beginning with what otherwise would 

 be the second section of it. Even with this peculiarity it is seen by 



