54 AET. I. B. HAYATA : 



Ilex formosana Maxim. ; Matslm. et Hayata I.e. p. 81. 



Hab. Uraisha, leg. N. Konishi, Aprili. 1908, (No. 78). 



Obseev. Branches blackish ; leaves elHptical acute at the base, 

 acuminate or cuspidate at the apex, obtuse at the very tip, 

 7-8 cm. long, 2-3 cm. broad, margin obscurely crenulate, veins not 

 very distinct on the upper surface, reticulated and dotted on the 

 under surface, somewhat pale beneath, petioles about 1 cm. long. 

 Flow^ers on a very short raceme which is almost contracted to a 

 cluster. 



Ilex goshiensis Hayata sp. nov. Eami validi, cinei.'ascentes, 

 ramulis fusco-rubescentibus angulatis. Folia alterna petiolata, 

 coriacea, obovato-oblonga, vel oblonga, 26mm. longa, 16 mm. lata, 

 apice retuso-acuta basi acuta, integerrima, supra opacissima, venis 

 non visis, subtus pallidissima tenuiter venosa, petiolis 4 mm. longis. 

 Drupae ad axillas foliorum fasciculatae, pedunculis 5 mm. longis, 

 pedicellis 3 mm. longis, globosse, 4 mm. in diametro, fusco-rubescen- 

 tes, basi sepalis 4 persistentibus instruetas. 



Hab. Shintiku : Goshizan, leg. Ï. Kawakami, (No. 1258.) 

 Near Ilex Championi Lœs. ; but differs from it in having, shortly 

 cuspidate leaves. The leaves of the other plant are very rounded 

 or even emarginate at the apex. This is also near Ilex memecylifolia 

 Champ., fi'om which it differs in having leaves which are retused at 

 the apex. There is in the Herbarium at Tokyo a specimen from 

 the Loo-choo islands wliich has been identified with /. 

 Hanceana Maxem., by Dr. T. Itö in the Tent. Fl. Lutch p. 367. 

 The specimen is very like, or even the same as, the plant 

 just described, and I have wondered if the Loo-chooan plant 

 be really identical with I. Hanceana. I have seen the 

 latter species at Kew and have noticed that it is differ- 



