144 The Philippine Journal of Science me 



both surfaces and slightly shining when dry, acuminate, base 

 acute or somewhat rounded, the margins very distantly and 

 obscurely denticulate or subentire, the middle leaflet longer than 

 the lateral ones, equilateral, 10 to 12 cm long, 2 to 2.5 cm wide, 

 its petiolule as long as the petiole, the lateral leaflets somewhat 

 inequilateral, sometimes slightly falcate, 7 to 10 cm long, their 

 petioles about 3 mm long; lateral nerves about 10, obscure, the 

 reticulations lax, subobsolete. Inflorescence pubescent, 2 to 3 cm 

 long, 2- or 3-pinnately umbellate-corymbose, the flowers umbel- 

 lately arranged on the ultimate branchlets, their pedicels 1.5 to 

 2 mm long. Female flowers: Calyx very obscurely lobed or 

 truncate. Petals 4, oblong, obtuse, 2 mm long, free, reflexed 

 in anthesis, pubescent externally, slightly thickened and 

 cucullate at the apex, often with an apical gland but not cor- 

 niculate. Ovary ovoid, pubescent, 1.5 mm long; stigma sessile, 

 4-lobed, the lobes stellate, narrowly oblong, 0.4 mm long. 



Luzon, Province of La Union, Bauang, Bur. Sci. 12964 Fenix, December, 

 1910, along streams, flowers pink. 



A species with quite the general appearance of Tetrastigma loheri 

 Gagnep., differing in its pubescent ovaries and stellate stigmas. 



TETRASTIGMA LOHERI Gagnep. in Not. Syst. 1 (1910) 265. 323. 



Tetrastigma philippinense Merr. in Philip. Journ. Sci. 7 (1912) Bot. 

 86. 



This species is merely enumerated to call attention to the reduction of 

 Tetrastigma philippinense Merr. to T. loheri Gagnep. Fragments of Loher 

 S71, the type of Gagnepain's species, received from Kew and from Dr. 

 Gagnepain, convince me that the form I described as Tetrastigma philip- 

 pinense under the impression that it was a species distinguishable from 

 T. loheri Gagnep. in certain floral characters, is really identical with 

 Tetrastigma loheri. It is accordingly here reduced. 



VITIS Linnaeus 



VITIS FLEXUOSA Thunb. in Trans. Linn. Soc. 2 (1793) 103; Planch, in 

 DC. Monog. Phan. 5 (1887) 347; Gagnep. in Mem. Soc. Soc. Hist. Nat. 

 Autun 24 (1911) reprint 28. 



Luzon, Benguet Subprovince, Baguio, Elmer 5880, March, 1904, with 

 perfect flowers and immature fruits, Williams 1052, June, 1904, with 

 mature fruits, For Bur. 5142 Curran, August, 1906, with mature fruits, 

 Bur. Sci. S371 Mearns, May, 1907, with male flowers, Bur. Sci. 13488 

 Ramos, May, 1911, with male flowers. 



Gagnepain, 1. c, has referred Elmer 5880 to the forma malaiana 

 Planch., previously known only from Java, but if the various forms desig- 

 nated by Planchon are to be maintained, there is no reason why Blume's 

 original name, sylvestris (sub Vitis), should not be retained. Gagnepain 

 notes that the different forms designated by Planchon pass into each other 



