10 BULLETIN 58, UNITED STATES NATIONAL MUSEUM. 



cially in the sources of the rapid Asahigawa or river of the rising sun. 

 For many years he collected the"hanzaki" in this part of the island, 

 and he mentions the localities of Tsuyama, Onaru Valley, Kuginuki 

 Valley, Mikamo, Maniwagori, all apparently in western Mimasaku. 

 To his very interesting article, Contributions to the Kjiowledge of the 

 Giant Salamander, in the Proceedings of the Department of Natural 

 History, Tokyo Imperial Museum, I, 1904, he has added a colored map 

 (Plate VIII) showing the distribution of the species. In this map 

 the extension westward to the provinces of Nagato and Suwo is 

 shown, and it also includes a large portion of the province of Idzumo. 

 The text affords, however, no opportunity of judging upon what ma- 

 terial this is based. This is particularly unfortunate as the testi- 

 mony as to the occurrence of this species so far west is quite con- 

 flicting. According to a memorandum kindly furnished me by Dr. 

 Hugh M. Smith, he was informed by Dr. T. Kitahara, zoologist of 

 the Imperial Japanese Fisheries Bureau in Tokyo, that the western 

 limit is the province of Bingo. The main drainage of the provinces 

 of Tamba, Tazima, and Iwami is toward the Sea of Japan, the others 

 drain to the inland sea, or to the Pacific, but the maps accessible to 

 me are not accurate and detailed enough to decide whether some 

 of the upper valleys of Iwami and Tazima do not in reality drain 

 southward to the Inland Sea. At any rate it can not be taken for 

 granted that the giant salamander occurs north of the watershed 

 until its occurrence in streams of the Sea of Japan drainage is proven 

 beyond a shadow of a doubt by actual specimens. 



In this connection it may be added that, according to Doctor 

 Smith's memorandum alluded to above, "Mr. I. Shishido, teacher of 

 zoology in the Third High School, Kyoto, states that the s})ecies is 

 reported from near Funatsa, province of Hida, in a river flowing into 

 the Japan Sea (Toyama Bay), but no specimens from that locality are 

 known to be in collections. The people in the vicinity of Funatsu, espe- 

 cially the women, eat this animal as a medicine." It is of the greatest 

 importance that this question, whether the giant salamander occurs 

 originally (not introduced) and regularly in the Japan Sea drainage, 

 should be definitely solved. The qualification "regularly" is added 

 because an isolated occurrence would not be convincing, since it might 

 be explained either as an accidental introduction, or as the result of a 

 Japan Sea stream having captured the sources of a Pacific Coast 

 stream. 



The capture of specimens in the Takadagawa, province of Musashi, 

 and at Kyoto, in 1832, in the canal of the castle Nizyono siro, as 

 related by Ilandangensei (Keisuke Ito.") probably, refers to animals 

 escaped from captivity or deliberately transplanted. 



aNihon Sanbutu, Sci. II, p. 39. 



