BETULACEAE. — CORYLUS 451 



(flowers); same region, R. Maack (with ripe fn its). Korea: Fusan, temple 

 gardens, September 6, 1903, C. S. Sargent (sterile). 



JAPAN. Hondo : prov. Kozuke, Mt. Asama, July 1904, U. Faurie (No. 5802; 

 with young fruits); prov. Shinano, Mt. Izuma, March 26, 1905, September 12, 

 1910 (ex Herb. Sakurai); prov. Suruga, district Tokyo, March 21, 1904 (and fall) 

 (flowers and old leaves, ex Herb. Sakurai). Kyushu: prov. Hizen, Nagasaki, 

 1862, R. Oldham (No. 746; sterile); without precise locahty, P. von Siebold (ex 

 Herb. Zuccarini sub nom. C . retusa; sterile); without locality and collector (ex 

 Herb. Lugd.-Bat. sub nom. var. Thunbergii; with flowers and young fruits). 



These specimens represent the typical form, which as far as I can judge from the 

 specimens and from Shirasawa's colored plate has yellow anthers and rather thin 

 leaves which are alwaj^s slightly pubescent on the rib and nerves of the under sur- 

 face, and never quite glabrous above. The shape is mostly rcctangular-obovate 

 with a more or less cordate base and a truncate and even emarginate apex bearing 

 a narrow acuminate middle lobe. In the Japanese specimens the leaves seem to be 

 a little more rounded at the apex, and further investigations are needed to show if 

 there are any other differences between the Siberian and northern Chinese forms 

 and those from Japan or not. I have not seen a Japanese specimen with such 

 ovate-cordate leaves as appear in Shirasawa's plate. 



In central China this species seems to be very variable, and there are probably 

 two varieties of which the one from the west may prove a distinct species on account 

 of its reddish anthers and other characters. As I point out on p. 446, the north- 

 western forms seem to have yellowish anthers like the typical form. The variety 

 from Yunnan apparently has dark anthers and is most nearly related to the forms 

 from Hupeh. For further remarks, see these varieties. 



Corylus heterophylla, var. sutchuenensis Franchet. See p. 445. 



Corylus heterophylla, var. jrunnanensis Franchet in Jour, de Bot. XHI. 198 

 (1899). — Burkill in Jour. Linn. Soc. XXVI. 504 (1899). — Winkler in Engler, 

 P/Zan2cwr. IV.-61, 48 (1904). — Schneider, III. Handb. Laubholzk. VI. 896 (1912). 



CHINA. Yunnan : " in sylvis prope collum Hee-chan-men, alt. 3000 m.," J. 

 Delnvay (type, ex Franchet) ; "in monte Tsang-chan," J. Delavay (No. 555, co-type, 

 ex Franchet); " Mengtsze, on way up to Mt. Benvonu," alt. 2000 m., September 

 13, A. Henry (No. 9682; shrub 0.6-0.3 m. tall; with ripe fruits); Mile district, 

 alt. 1800 m., November 1, A. Henry (No. 9894; shrub, 1.8 m. tall; old leaves and 

 catkins). 



I have not seen the tjqae, but Henry's specimens agree well with Franchet's 

 description. This variety is distinguished from the others by its leaves being thickly 

 tomentose-hirsute on the whole under surface, by its more tomentose and glandular 

 branchlets and petioles which are short and thick, and by its more distinctly velu- 

 tinous nuts, but there are more or less intermediate forms apparently connecting 

 this southern variety with that of central China. The color of the anthers seems to 

 be red. 



7. Corylus hallaisanensis Nakai in Fedde, Rep. Spec. Nov. XIII. 250 (1914); 

 in Tokyo Bot. Mag. XXIX. 36 (19 J5). 



NORTHEASTERN ASIA. Korea: Quelpaert, Hallai-san, forests, August 

 1909, Taquel (No. 3239; sterile); same locality, October 1909, Tagtiet (No. 3238; 

 sterile); same locality, September 1903, raqmt (No. 333, tyjie ex Nakai). 



Taquet's specimens of 1909 are from the type locality and agree with Nakai's 

 description, but unfortunately I have not seen the fruits which are described by 

 Nakai (1914) as follows: " Perigonium fructifenim hinsutum compacte clausum 

 ovatum apice mucronatum lobulatum, cum mucrone 3 mm. longo 18 mm. longum 



