222 michelson: American Indian languages 



to exaggerate the importance of insect transmission, and to overlook, 

 even in cases where insects may occasionally be concerned, the greater 

 importance of other modes of infection. This is indicated by Sambon's 

 theory of transmission of pellagra by Simulium — a theory which was 

 advanced with enthusiasm on the ground that it fitted into the known 

 facts in the epidemiology of the disease. It took two years of hard 

 work on the part of members of the force of the Bureau of Entomology, 

 working in collaboration with the Thompson-McFadden Pellagra 

 Commission, to upset this theory in a thoroughly scientific manner. 

 As has been pointed out several times of late, there is always consider- 

 able danger in' conclusions based on epidemiological findings. Trans- 

 mission experiments are necessary. 



One conclusion must be drawn which can hardly be disputed: There 

 is an enormous field for the entomologist in the careful study of all 

 of the aspects of the biology of not only those insects which have 

 already been shown to be disease carriers but of those which are likely 

 to be implicated. It is to the trained economic entomologist that we 

 must look for the methods of destruction of these insect carriers, and 

 the prevention of this class of diseases lies at his door rather than at 

 that of the physician. Either that, or sanitarians must be trained in 

 what is now known as medical entomology. 



ETHNOLOGY. — Remarks on American Indian languages, a 

 study in method? Truman Michelson, Bureau of Ameri- 

 can Ethnology. 



At the very beginning of this subject it should be stated that 

 there is no single type of speech which holds good for all Ameri- 

 can Indian languages. The statement that all American Indian 

 languages are both polysynthetic and incorporative, so con- 

 fidently affirmed by the older writers, is false. The number of 

 American Indian languages that are either polysynthetic or 

 incorporative, is extremely limited indeed. I do not know of a 

 single feature that may be said to be characteristic of all American 

 Indian languages. Even so, a combination of certain features 

 is quite sufficient to determine whether any given language is an 

 American Indian language or not. It is this which enables us 

 to say without any hesitation that Chuckchee, Koryak, and 

 Yukaghir (which are spoken inmortheastern Asia) are American- 

 oid languages. If they were spoken in America we would call 

 them American Indian languages. They do not belong geneti- 



1 Printed with the permission of the Secretary of the Smithsonian Institution. 



