538 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1921. 



always burns it on leaving it, lest his soul, having taken a fancy to 

 it, should stray back there by itself. 



To return to lycanthropy in particular, the practice described, as 

 distinct from the belief, seems particularly associated in Assam with 

 the immigration from the northwest — that is, from the direction 

 of Nepal and Thibet. The Changs probably have an admixture of 

 Singpho blood, and the Singphos are known to have come from that 

 direction; so, too, the Kacharis who, like the Changs, have a clan 

 of tiger men, and call it the Mosa-aroi, and the Meches w T ho have a 

 corresponding clan called Masha-aroi, which also goes into mourn- 

 ing for the death of a tiger — both came from the north of the Brah- 

 maputra. Among the Garos also the practice is found, and they 

 too came from the same direction. On the other hand the Khasis, 

 who seem to belong to a different stock — perhaps to the Kol-Mon- 

 Annam race, and to have come from the east — say they believe in 

 the existence of tiger men, but appear to have absorbed the idea 

 from the Garos, who are their neighbors, and not to have possessed 

 it as an indigenous idea, nor to indulge in, or believe that they in- 

 dulge in, the practice themselves. The Angami, who also does not 

 practise lycanthropy, again seem to have immigrated into the Naga 

 Hills from the southeast and to be intimately connected with the 

 Bontoc Igorot of Luzon in the Philippines. In other ways, how- 

 ever, particularly in language, the Sema is connected with the 

 Angami, though on the other hand there are points of culture which 

 keep suggesting a connection between the Sema and the Garo. One 

 of them is the use of Y-shaped posts to celebrate feasts given to the 

 village, similar wooden posts being used by the Garo, though he is 

 at present entirely isolated from the Sema, while the Kachari ruins 

 at Dimapur contain the same bifurcated monuments in stone. Per- 

 haps the explanation is that the present Sema tribe is the result of 

 the amalgamation of a small Angami element which has imposed 

 itself upon another stock, a process which the Sema tribe itself is 

 still carrying on with regard to its neighbors to the east at a very 

 rapid rate, a Sema chief or adventurer grafting himself and a few 

 followers on to a Sangtam or Yachungr village ; this in a generation 

 or less becomes entirely Sema in language and polity, though no 

 doubt retaining its former beliefs and certainly retaining much of 

 its former ceremonial. 



The theory that this form of lycanthropy comes from a northern 

 source is perhaps supported by the fact that the form which the 

 belief takes in Burma and Malay, as well as in the plains of India, 

 seems to turn on an actual metamorphosis of the body. Mr. Grant- 

 Brown, writing in the Royal Anthropological Institute's Journal in 

 1911 about the Tamans, a tribe of Chinese origin in the Upper Chind- 



