4 BUREAU OF AMERICAN ETHNOLOGY [Bull. 90 



differ in the two versions are listed below, followed by an explanation 

 of each fact:^^ 



Ay. W. 



Line 1. yu'a-.a'i (written twice) Line 1. yuV.a'i (written three times) 



(1) 



Line 2, 8. g'-^Dale-Gwo'Ii. Line 3, 9. g-'^Dali e-Gwo'M (2) 



Line 4. dunu'-y'tam'le-.i^ Line 5. dunu'-y'tamle"' (3) 



Line 12. n()-''dadu'-gta'Q-°sti' Line 12. nQ-^tadu'-kta'g-'-sti' (4) 



Line 14. de'-du-dg-ne-li'se-sti' Line 14. de'-du'do-ne-lidi'se-sti' (5) 



Line 14. g9-"tsa'd(o)tagi-ya' Line 14. gQ-"tsa'tagfya' (6) 



Line 16. widistatl(i)tadinQtaniga Line 16. widistotl(a)tadi ... (7) 



Line 20. atsflo"' Line 21. atsfla' (8) 



(1) Whereas Ay. has written the song-word twice, W. writes it 

 three times; neither of them is right, since, strictly speaking, it should 

 be written seven times; but it is very rare that this is done; often we 

 even find these song-words written only once, since every medicine 

 man knows that they are to be repeated four or seven times anyhow. 



(2) It is customary for the Cherokee who write a great deal in 

 the Sequoya syllabaiy to adhere to a "sandhi"-law of the spoken 

 language, and to drop a final vowel before a word beginning with a 

 vowel, linking the consonant of the first word with the vowel of the 

 second as in this case: (9'Da) li4-€*(gwo)>-le"- 



It will be noticed that Ay. conforms to this use in every one of the 

 three cases where the word occurs (Ay. Unes 2, 8, 16), whereas W. 

 does it only in the last case (W. line 16). This discrepancy is to be 

 explained by the fact that I read out the text in slow tempo, and by 

 so doing no "sandhi" phenomenon was heard by my informant. 



(3) In the written as well as in the spoken language the -i, at the 

 end of the -le*i, -ne-i, -se-i and similar tense-suffixes is written and 

 pronounced when the sentence is considered as finished; if more 

 words follow in the sentence, however, it is generally dropped. It 

 is a mere matter of eiiphony, to which W. has in this case not con- 

 formed, probably because I may have led him to believe by the 

 intonation of my voice that the sentence was not finished. 



(4) In the Cherokee syllabary the system of the surd and sonant 

 velars and dentals is very imperfectly worked out. As a result, the 

 Cherokee themselves are quite inconsistent in using the symbols for 

 g, k and d, t. The matter is made more complicated by the actual 

 existence of the so-called "intermediates" in their phonetics. This 

 discrepancy is an illustration of this state of affairs. 



(5) Ay. omitted the symbol for the -di- syllable here, without 

 which the word has no meaning. W. consequently interpolated it. 



(6) Although such phonetic phenomena as breath, stops, etc., are 

 quite frequent in Cherokee linguistics, the syllabary very imper- 



1' The figures in parentheses following the words as written by W. refer to the 

 explanations in the following paragraphs. 



