INFLECTION FOR CASE. 467 



distributively. The inflection of the adjective and numeral adjective is not 

 quite so rich in forms as that of the substantive. 



CASE-SUFFIXES AND CASE-POSTPOSITIONS. 



The numerous forms of nominal inflection, called cases by gramma- 

 rians, may be divided into two categories: («) the \mre\y ffrnmmatic cases, 

 expressing mere relation of one noun to another, and being only three in 

 number, the subjective, direct-objective, and possessive case; (b) all the other 

 cases, as instrumental, inessive, adessive. They are either locatives or take 

 their origin in some locative relation of the noun to the verb. 



But this purely logical division of cases does not always work well 

 when practically applied to existing languages. It cannot be rigidly ap- 

 plied in a grammar of the Klamath language, for here the case of the 

 direct object is also that of the indirect object, and the possessive case is 

 also that which corresponds to the Latin ablative when connected with a, 

 db and a verb in the passive voice. 



It is best to divide the cases of Klamath inflection into cases formed 

 by case-suffixes and cases formed by case-postpositions. The former I call, 

 for convenience, suffix-cases, the latter postposition-cases. 



Suffix-cases are formed by nominal inflectional suffixes having no inde- 

 pendent meaning for themselves as ivords. They are usually unaccented, 

 and only two of them are dissyllabic in their unabridged form. Besides 

 the subjective case, which is not always made distinct by a suffix, there are 

 the cases in -ash, -am (-lam), -ti, -tka, -tat, -xem, -na, -emi. 



Postposition-cases are formed by means of particles having an inde- 

 pendent signification for themselves as ivords of the language, but when 

 connected with a noun are never placed before it. They are all of a loca- 

 tive import, and frequently take the accent. Their list is: -i, -kshi, -ksaksi, 

 -tala, -tana. 



The function of some of the case-suffixes is of a very general nature, 

 and should be illustrated by a large array of quotations to be made clear. 

 No noun of both dialects will be found which is in possession of all the 

 case-suffixes and case-postpositions, and of the fourteen found in the para- 

 digms hardly ten can be said to be in constant use. Temporal suffixes, for 



