50 REPORT OF THE SECRETARY. 



of 1900-1901, with acc()iiij)anyinii translations. Subsequently he was engaged in 

 copying and translating a secontl set of texts obtained at Masset, Queen Charlotte 

 Islands, during the same expedition. 



On December 16 Doctor Swan ton returned to the Northwest to engage in field work, 

 particularly among the Tlingit Indians of the Alaskan coast. From January 9 to 

 March 21 he was at Sitka among the northern Tlingit, and from then until May 5 

 among the southern members of that family. On the way thither he engaged in 

 some incidental work among the Haida, and during the season he collected about 

 one hundred Tlingit myths, as well as much ethnologic material in other branches. 

 One of Doctor Swanton's main objects in this expedition was to define the relations 

 between the Haida and Tlingit peoples, looking to the possibility of a genetic rela- 

 tionship between them. A final conclusion on this point can not yet be given, but 

 it was discovered that many of the resemblances noted between the two languages 

 are due to an early residence of the Tlingit opposite the Haida on the coast now occu- 

 pied by the Tsirashian. This fact, already partially recognized, and now practically 

 demonstrated by Doctor Swanton, results in limiting the origin of much of the culture 

 on this coast to the immediate neighborhood of Hecate Strait, northern British 

 Columbia. An important contribution to the general subject of clan organization 

 was made by the discovery of a small Tlingit group which practiced marriage with 

 either of the two great clans. 



Since his return to the office Doctor Swanton has been engaged, first, in revising 

 the Tlingit material for the Cyclopedia of Indian Tribes in the light of the fresh 

 information gained during his recent trip; and, second, in copying the texts taken 

 among the Tlingit. 



Mr. H. H. St. Clair, 2nd, special assistant in Philology, visited northern California 

 and southern Oregon early in the year for the purpose of collecting data among the 

 Rogue River, Coos, and other small tribal remnants, of which a few individuals sur- 

 vive in that region, and a number of valuable vocabularies were secured. 



OFFICE WORK. 



Mr. J. N. B. Hewitt, ethnologist, has continued the preparation and proof-reading 

 of part 1 of his monograph onlroquoian cosmology, which is to appear in the Twenty- 

 first Annual Report. The reading of the galleys of the interlinear translation and 

 the free translations into English of the Onondaga, the Seneca, and the Mohawk 

 versions of this cosmology occupied a large part of his time during the year. Exten- 

 sive revision was required on account of the premature transmission of the manuscript 

 to the ])rinter in 1902. This caused much delay, but it was considered advisable 

 to permit the delay rather than to have the paper ])ublished in a form unsatisfactory 

 to the author. 



As custodian of linguistic manuscripts, Mr. Hewitt, assisted by Miss Sniedes, con- 

 tinued the work of revising and bringing up to date the card catalogue of the lin- 

 guistic and other manuscripts in the archives of the Bureau. This card catalogue 

 was originally prepared in 1896-97 by Mr. Hewitt, with the assistance of the late 

 Rev. J. Owen Dorsey, and in this work the manuscripts were classified under three 

 main heads: First, the author or collector; second, the tribe or band-village; and 

 third, linguistic stock — all under one alphabet. The cross-reference catalogue f)f 

 the names of tribes and villages noted in the manuscripts is of very great use for 

 the purpose of comparison and research and for determining the number and distri- 

 bution of these manuscripts among the various stocks. In the i)resent revision 

 the work consists in making a dui)licate coi)y of the card descriptive of the manu- 

 script, which dui)licate is ])asted on tlie jacket or package containing the manuscript. 

 A nundjer is affixed to the original card, to the duplicate, and to the manuscript 

 itself for the purpose of ready identification. New cards are being made for manu- 

 scripts acquired since the completion of the original catalogue in 1807. This scheme 



