INTRODUCTION. 203 



less than in their morphology. Generally the structure of a sentence is 

 simple, being based only on the coordinative and adversative principle. But 

 where there is a lack of the relative pronoun, or an inadequate supply of 

 conjunctions, as in the dialects of the Mask6ki family, verbals are necessary 

 to supplant them. This produces encapsulated sentences, which, by the 

 frequent repetition of the verbal, soon become tiresome through monotony, 

 and diminish the perspicuity and compreheusibility of the spoken word. 



A continued study of the Klamath language has convinced me that it 

 occupies a middle position between the extremes of synthetic and analytic 

 structure just referred to, but that, nevertheless, it shows very plainly all 

 the characteristics of agglutinative tongues. The distinction between the 

 noun and the verb is made pretty clear, although most substantives can be 

 considered as nomina verhalia; the verb is not overloaded with forms point- 

 ing to material ideas, neither with tenses, modes, nor voices, and possesses no 

 real personal conjugation. As to derivation, Klamath is undoubtedly poly- 

 synthetic in its affixes, the suffixes preponderating largely over the prefixes, 

 and differing from them in their functions. Outside of Klamath and the 

 dialects of the Dakota stock, but few languages have been discovered in 

 which the prefix indicates the exterior form of the verbal subject or object, 

 or even the quality of the verbal action. Reduplication for inflectional 

 purposes is as well developed hei-e as it is in Pima and Selish and forms 

 one of the characteristic features of the language. As to its syntax, Kla- 

 math may be called analytic; a profusion of conjunctions relieve it of the 

 too frequent use of participial and similar constructions, as does also the 

 relative pronoun kat, and the use of the substantive verb gi simplifies the 

 verbal inflection to a great extent. 



These and other characteristics impart to the language of the Md- 

 klaks a well-defined type, and approach it to some of the tongues of 

 modern Europe, in which analysis has not preponderated over synthesis. 

 An attentive study of the numerous texts obtained from the Indians, paired 

 with constant comparison of Klamath structure with the structure of many 

 foreign and American languages, could alone furnish a solid basis for 

 establishing the grammatic rules of this upland tongue. The rhythmic, 

 stately, and energetic tenor of its periods, especially those of the larger 



