58 BULLETIN 58, UNITED STATES NATIONAL MUSEUM. 



of web, on the contrary show a minimum size of the interdi(2;ital 

 membranes. 



A very similar form, possibly confined to Shikoku, having the 

 deeply excised webs, long parotoids, long limbs, and slender digits, is 

 very near to B.formosus, but has a much smaller tympanum and first 

 and second fuigers subequal. 



The third form is characterized by small tympanum, stouter and 

 shorter limbs, much larger webs, the true B. ja'ponicus. 



In spite of the general similarity in coloration and otherwise it is 

 difficult to escape the conclusion that the Japanese toads represent 

 three more or less distinct forms, and as it is at present impossible to 

 establish their interrelationship clearly it is thought best to designate 

 them by binominal appellations. The great difficulty arises in 

 ascribing definite geographical limits to them. Unfortunately, a 

 large number of the specimens examined are without indication of 

 their exact origin. Thus all the specimens in the Leiden Museum 

 collected by Buerger and von Siebold are only labeled '' Japon." Sev- 

 eral in the British Museum are similarly deficient. The only true 

 B. japonicus with a definite locality attached seems to be one from 

 Kobe in the British Museum and one from Miyakoshima in Science 

 College Museum. On the other hand, it would almost appear as if all 

 the B.formosus of which the exact locality are known came from near 

 Tok3"o and Yokohama (Camerano's specimens; types in British 

 Museum; specimens in United States National Museum) and 

 Dagoshima, wherever that locality may be. <^ We have thus an indi- 

 cation of the B.formosus being eastern and northern, while the one 

 with the larger webs, etc., may be more western. 



Finall}', the only authentic locality from which I have seen speci- 

 mens of the new form, further on described as B. sniithi, is the island 

 of Shikoku. There is a specimen in the Imperial Museum in Ueno 

 Park, Tokyo, apparently belonging to this form, which w^as in a bottle 

 with the type specimen of Polypedates isliihawse and said' to have come 

 from Okinawa shima, Riu Kiu. As no toad has been recorded by any 

 collector from that archipelago, the locality given becomes very 

 doubtful. 



The next question then is whether any one of the forms wliich we 

 thus recognize as occurring in Japan is identical with the toads occur- 

 ring on the mamland. I have already quoted Boulenger to the effect 

 that there are several forms, though he does not recognize them by 

 name because they intergrade. Let it be understood at once that the 

 form m Japan most remote from the mainland forms is B. formosus. 

 It is consequently the so-called B. vulgaris japonicus, i. e., the one 

 with the greater webs, shorter parotoids, shorter legs, etc., which 



a If Dogashiraa is meant, as surmised by me. the locality agrees well with the 

 others. 



