646 ANTHROPOLOGICAL PAPERS. 



There bad been considerable discussion as to whether words which 

 sound alike show a similarity of origin in the tribes speaking them j 

 but Prof. W. 1). Whitney says that numerals and words indicating re- 

 lationship are of more value as evidence on this point than any others. 

 A further comparison of the four languages which accompany this shows 

 that the Twana and Nisqually agree in many points, and the Klallam 

 is similar to them in some, but the Chemakum, except in a few instances, 

 is different from all the others, and these instances are such that the 

 words may lifive been adopted from the Klallam, their neighbors. Dr. 

 Gii)bs* is doubtful whether the Chemaknms ought to be included in the 

 Selish family or not. There is no doubt but the}' are connected with 

 theHohs and Kwilliuts, judging from the numerals and their traditions.t 

 Dr. Gibbs is also of this opinion. 



In this connection I would say that a comi^arison of the manners and 

 customs of all the above-named tribes, as far as I have been able to 

 learn them, show that they are much. the same, varying only as their 

 different environments compel them. 



Whether or not similarity of customs shows a similarity of origin 

 seems to be an open question. If anj customs do, it is to my mind their 

 religious rites, for experience and observation prove that they itre less 

 likely to change these than any other. In customs there is little differ- 

 ence among these tribes, the more northern Indians being a little more 

 savage. 



The Twana language has another form of numerals for counting 

 money, stones, and small round things, which differs from the form 

 given mainly in adding to it lis, elis, alis, or ta lis as a final termina- 

 tion. The Skwaksin a second series, used mostly for counting money, 

 which is made by adding els. The Klallam has also a second form, used 

 in the same way as the Twana and formed by annexing ut-hu or aiut- 

 hu; and also a third form for counting animals, which annexes to the 

 original eks, uks, or e iks. In counting large numbers they cut notches 

 on sticks to assist their memories. I am not aware that any of them 

 knew anything about multiplication before the whites came, except by 

 repeated additions. 



Time. — The year was divided into thirteen moons, for each of which 

 they had names, but I have not been able to learn that they had any 

 names for particular days. The day-time was divided into dawn, sun- 

 rise, forenoon, noon, afternoon, sunset, and dusk, nnd the night had 

 only the division of midnight. The present names of the days of the 

 week have been adopted since the whites came among the Twanas. 

 They signify holy day, one day past, two days past, etc. 



Length. — They had four standards of measurement of length : (1) from 

 end to end of the middle finger outstretched; (2) from the shoulder to 

 the end of the corresponding hand, arm extended ; (3) from the shoulder 



' Vol. 1 Contributions to American Ethnology, pp. 159-177. 

 tSeo Swan's Indians of Cape Flattery, p. 57. 



