234 BUREAU OF AMERICAN ETHNOLOGY [bull. 40 



derived, from the same stem, but employ the vocative form of mother 

 and FATHER respectively. Of other vocatives, Tdu'yam^ o friend! 

 31.6, 8; 32.4, 6 is the bare stem; liaml 70.5; 71.7, the stem with 

 added characteristic -i- ; liinde o mother! daughter! 56.7; 76.10, 

 13; 186.14 is quite peculiar in that it makes use of the first personal 

 singular ending irde) peculiar to nouns with possessive suffix and pre- 

 ceding pre-positive. Only two other instances of a nominal use of 

 -de without pre-positive or local adverb have been found: mo'fe^ my 

 son-in-law! (as vocative) 164.19; and ¥wi'naxde my folks, rela- 

 tions, which otherwise follows Scheme II (e. g., third person 

 ¥wi'naxdd"^) . 



The normal pronominal suffix of the third person is -xa; -a is found 

 in only four cases, Icla'sa his maternal grandparent, lia'sa his 

 MATERNAL UNCLE, fa'da HIS PATERNAL AUNT, and Jia'sda his brother- 

 in-law. The first two of these can be readily explained as assimi- 

 lated from *Tc!a'sxa and ^Tia'sxa (see § 20, 3); *t'adxa and *hasdxa, 

 however, should have become ^fa'sa and ^ha'sa respectively. The 

 analogy of the first two, which were felt to be equivalent to 

 stem + -a, on the one side, and that of the related forms in -d- 

 (e. g., fadd and Jiasdd) on the other, made it possible for fa' da and 

 Tia'sda to replace *t'a'sa and *Jia'sa, the more so that a necessary 

 distinction in form was thus preserved between Jia'sa his maternal 

 uncle and Tia'sda (instead of ^ha'sa) his brother-in-law. 



The difference in signification between the third personal forms in 

 -xa and -xagwa (similarly for the other pronominal schemes) will be 

 readily understood from what has already been said, and need not be 

 enlarged upon: 



ma'xa wd°^-MmiY he spoke to his (some one else's) father 

 ma'xagwa wd^-MmiY he spoke to his own father 



There is small doubt that this -gwa is identical with the indirect 

 reflexive -gwa of transitive verbs with incorporated object. Forms 

 in -gwan seem to refer to the plurality of either possessor or object 

 possessed: 



Tc' aba' xagwan their own son or his (her) own sons 



elxdagwan their own canoe or his own canoes 

 The final -n of these forms is the indefinite plural -an discussed 

 below (§ 99). Plural (?) -gwan is found also in verb forms (144.12; 

 150.24). 



1 Ic.'uyam- is perhaps derived, by derivational suffix -{a)m, from verb-stem ktony- go together with one. 



§ 91 



