1886.] 41 i [Gatschet. 



The use of the Roman letters with the value they have in the English 

 alphabet. This alphabet is wholly preposterous, even for English itself, 

 and much more so for any foreign, especially illiterate languages. If the 

 authors had been more accurate in their transcription of the words re- 

 ceived, they would not have used ch sometimes for /, at other times for 

 tch ; cf. the numerals 2, 13. 



Instances where the authors failed to hear sounds with sufficient accu- 

 racy ; cf. cattle (p. 421). 



Insufficiency of the knowledge of English on the side of the two female 

 informants; cf. the mistaking of wet for lohite. It appears that several 

 terms were obtained not by putting questions, but by making gestures ; in 

 many vocabularies of other languages this has become a fruitful source of 

 errors. Compare the term obtained for islands with that for ship, vessel 

 (mamashee), and mouth with tongue. 



The want of distinction between the noun and verb in English often 

 causes grammatic confusion, as in the case of lead, sleep, scratch, etc. 

 Moreover, the verb is sometimes placed in the participle, sometimes in the 

 infinitive, especially in Rev. Leigh's vocabulary. 



A few other remarks referring to the present condition of the vocabu- 

 laries are as follows : 



In several terms the initial sound has been dropped, either through in- 

 accurate hearing or incompetency of the copyists : osweet for kosweet 

 deer, ewis for kewis watch, cf. also obosheen with boobasha, oosuck with 

 woas-sut, eesheet with mamesheet. 



Instances of contraction by synizesis, ellipsis, etc., are not unfrequent : 

 a'shoging from ashwoging arrow ; bedoret from bogodoret heart ; shuco- 

 dimit from shucododimet "Indian cup." 



The month-names were obtained by Cormack and are partly misspelt 

 and faulty. It is very doubtful to me that April, June and September 

 were all called by the same term, the two final syllables of which contain 

 the word yaseek one, perhaps signifying one and first. But in American 

 languages two successive Indian moons are often observed to possess the 

 same name, as we see it done here in the case of October and November, 

 whose names coincide pretty closely. 



GRAMMATIC ELEMENTS. 



Phonetics. 

 The points deducible with some degree of certainty from the very im- 

 perfect material on hand may be summed up as follows, the sounds being 

 represented in my own scientific alphabet, in which all vowels have the 

 European continental value : 



Vowels : a a 



e a o 



i I u ii 



Diphthongs : ai, ei in by-yesh birch, madyrut hiccough ; oi in moisamad- 

 rook wolf; ou, au in ge-oun chin; oe may indicate 6 : emoethook (?), etc. 



PROC. AMER. PHILOS. SOC. XXIII. 123. 3A. PRINTED JUNE 12, 1886. 



