PRIMEVAL JAPANESE. 801 



The progress of civilization is always gradual. A nation does not 

 pass, in one stride, from })urial in rude tumuli to sepulture in liighh' 

 specialized forms of stone vaults, rior 3'et from a l)ronze age to an 

 iron. It is therefore evident that the evolution of dolmen from bar- 

 row did not take place within Japan. The dolmen constructor must 

 have completely emerged from the bronze age and abandoned the 

 fashion of barrow burial before he reached Japan. Otherwise search 

 would certainly disclose some transitional form between the barrow 

 and the dolmen, and some iron implements would occur in the bar- 

 rows or bronze weapons in the dolmens. If, then, the barrow builder 

 and the dolmen constructor were raciall}' identical, it would seem to 

 follow that the latter succeeded the former by a long interval in the 

 order of immigration and brought with him a greatl}^ improved t} pe 

 of civilization evolved in the country of his origin. 



The reader will he naturally disposed to anticipate that the gecr- 

 graphical distribution of the dolmens and the barrows furnishes some 

 aid in solving this problem. But though the exceptional number 

 found on the coasts opposite to Korea tends to support the theory that 

 the stream of Mongoloid immigration came chiefly from the Korean 

 Peninsula via the island of Tsushima, there is not any local difl'erenti- 

 ation of one kind of sepulture from the other, and, for the rest, the 

 grouping of the dolmens supplies no information except that their 

 builders occupied the tract of country from the shores opposite Korea 

 on the west to Musashi and the south of Shimotsuke on the east, and 

 did not penetrate to the extreme northeast or to the regions of moun- 

 tain and forest in the interior. 



Here another point suggests itself. If the fashion of the Japanese 

 dolmen was introduced from abroad, evidences of its prototype should 

 survive on the adjacent continent of Asia. If the numerous dolmens 

 found on the coasts of Kiushiu and Isumo facing Korea are to be taken 

 as indications that their constructors emigrated originally from the 

 Korean Peninsula, then Korea also should contain similar dolmens, and 

 if an ethnological connection existed between Japan and (Jhina in pre- 

 historic da3\s, China, too, shoidd have dolmens. But no dolmens have 

 hitherto been found in China, and the dolmens of Korea differ radi- 

 cally from those of Japan, being "merely cists with megalithic cap- 

 stones"' (Gowland). It has been shown, further, that dolmens similar 

 to those of Japan are not to be found in an}' part of continental Asia 

 eastward of the shores of the Caspian Sea, and that western Europe 

 alone offers exacth' analogous types. In short, from an ethnological 

 point of view, the dolmens of Japan are as perplexing as the dolmens 

 of Europe, and the prospect of solving the riddle seems to be equally 

 remote in T)oth cases. All that can be affirmed is that the dolmens 

 offer strong corroborative testimony to the truth of the Japanese his- 

 torical narrative which represents Jimmu as the leader of the last and 



