SAXIFRAGES OF YEZO AND THE KURILE ISLANDS 111 



by the same authority as S. cortusifolia. The specimen I 

 examined at Kew Herbarium is not *S'. cortusifolia at all, but 

 represents S. nipponica Makino (c/. Tokyo Bot. Mag. xxiv. p. 1). 

 According to Maximowicz, Siebold and Zuccarini's S. cortusifolia 

 includes two plants, one of which is S. madicla [S. cortusifolia var. 

 madida Maxim.) Makino ; their description "... petalis unguicu- 

 latis, duobus quam reliqua tria ovato-spathulata pluries longioribus 

 linearibus acuminatis ..." suggests S. madida. I do not venture 

 to divide S. cortusifolia Sieb. & Zucc, and employ the name for 

 the plant fairly well characterized and figured by Maximowicz 

 (I.e.) and Yatabe {I. c). 



3. S. MADIDA Makino in Tokyo Bot. Mag. xv. p. 11 (1901). 

 Syn. S. cortusifolia (i madida Maxim. I.e. Yatabe, I.e. t. vii. 

 Nom. Japon. Jinjiso. 



Hab. Yezo : " sommet de Kishiri " {U. Faurie ! 1892, no. 8397 

 pro parte). 



Obs. Faurie's collection no. 8397 is represented by two species 

 of the genus. The one is S. puntata L. (cf. infra), the other is 

 the present species, which Boissieu {I. c.) took for S. cortusifolia. 

 It is difficult to understand the confounding of these well-marked 

 plants. I agree with Makino's separation of this plant from 

 S. cortusifolia. 



The present species is distributed mainly in Southern and 

 Central Japan, where the northern limit of localities is about 

 37° N. ; it is thus highly interesting to find it in Hokkaido. The 

 above-mentioned specimens include two forms — one normal, the 

 other with more or less incised lower petals, which I call forma 

 incisa {cf. Tokyo Bot. Mag. xxiv. p. 64 (1910) ). 



4. S. PUNCTATA Linn. Sp. PI. p. 401. Engl. /. c. p. 137. 

 Boiss. I. c. p. 686. Yabe et Yendo in Tokyo Bot. Mag. xviii. 

 p. 184 (1904). 



Syn. S. cEstivalis Fisch. Fr. Schm. Reis. Amurl. Sachal. p. 133. 



Nom. Japan. Chishima-iwabuki. 



Hab. Yezo : " sommet de Rishiri" {U. Faurie ! 1892, no. 8397 

 ex parte ; W. Hirosel 1896). Kurile : insula Shimushir ; in port. 

 Broton (iTof/ama ! 1893); ins. Shimushu (SeZci ! 1895; M.Aizaival 

 1900; K. Yendo \ 1903). 



Obs. The distribution of this plant in Japan is restricted to 

 Hokkaido and the island of Saghalien. 



5. S. FuscA Maxim, in Mel. Biol. viii. p. 602. Fr. et Sav. I. c. 

 p. 146. Miyabe, Fl. Kuril, p. 233. Boiss. I. e. p. 686. 



Nom. Japon. Kurokumoso. 



Hab. Kurile : ins. Eturup : sine loco speciali ( U. Faurie ! 

 1891, fr.) ; " vicinity of Tsurubetsu " {K. Miyabe ! 1884) ; Porosu 

 {T. Kawakami ! 1898) ; Mt. Chirip {T. Kawakavii ! 1898). 



(B divaricata Fr. et Sav. I. c. ii. p. 354. 



Nom. Japon. Yezokurokumoso. 



Hab. Yezo : sine loco speciali {Maries !) ; Osatsube, prov. 

 0^h\mi{K. Miyabe; 1890); Iwanai (r«/cc??o6?i! 1883); Wakkaoi- 

 sando {K. Miyabe ! 1884) ; ad fl. Hacham superiorem, prope 

 Satporo (//. Takcdal 1908); Yuparo {Y. Tokuhuchi \ 1893); 



