Araliaceae. 



Tetraplasandra meiandra (Hbd.) Harms. 

 (Plates 137, 138, 139.) 



TETRAPLASANDRA MEIANDRA (Hbd.) Harms in Eng. et Prantl Pflzfam. Ill, 8 

 (1898) 30. Triplasandra meiandra Hbd. Fl. Haw. Isl. (1888) 152; Del Cast. 111. 

 Fl. Ins. Mar. Pacif. VI (1890) 184. Triplasandra Waimeae (Wawra) Heller 

 PI. Haw. Isl. (1897) 871. Heptapleurum (?) Waimeae Wawra in Flora (1873) 

 158. (Wawra's specific name Waimeae should hold good on account of priority, 

 but as there is already a species with that name in Tetraplasandra, Hillebrand 's 

 moiandra is here adhered to.) 



Leaves 3 to 4.5 cm long, the petiole dilated at the base and clasping; leaflets 7 to 

 J3; inflorescence, umbellate but variable: either the pedicels at the end of 3 to 5 ter- 

 minal peduncles (simply umbellate, but then shrubs), or at the ends of rays which 

 proceed from the ends of 3 to 5 peduncles, the latter rarely united by a common rhachis 

 (compoundly umbellate); bracts broadly ovate, 4 to 8 mm long, caducous long before 

 the flowers expand; calyx cylindrical, ovate or obovate, with a short denticulate or 

 undulate border; petals 5 to 8, triangular or linear lanceolate; stamens as many, shorter, 

 or as long as the petals (in one variety only). Ovary 2 to 6 celled; stigmas 2 to 6, 

 sessile on the conical vertex, or, when 4 to 6, raised on a short stylopod, drupe cylindrical, 

 ovate, oblong, or obovate, or subglobose. 



Hillebrand, in his Flora of the Hawaiian Islands, says in a foot note on 

 page 152 : 



"Under this collective species I have united the following forms, which are 

 exceedingly rare, each corresponding to a single or a few individuals, found in 

 closely-circumscribed localities. ' ' 



He then describes six varieties, as follows : 

 Stigmas 2, rarely 3. 



a. 7 to 12 leaflets. 

 /3. 7 to 9 leaflets, 

 y. 11 leaflets. 

 Stigmas 3 (4). 



8. 7 to 9 leaflets. 

 Stigmas 3-4-5. 



e. 9 to 13 leaflets. 

 Stigmas 4-5-6. 



. Leaflets as in 8, drupe ovoid. 



The above key to the varieties of this species can not be relied upon, as one 

 may find plants with only 2 stigmas and 13 leaflets, and plants with 4 stigmas, 

 6 stamens and 9 leaflets. The specimens from Oahu are more or less shrubs, and 

 have rather long rays or peduncles, while the plants from the other islands are 

 always trees, and have rather short rays. It is unfortunate that Hillebrand did 

 not define them more clearly. However, complete material is not always possible 

 to obtain, and therefore an exact diagnosis not always possible, as the stamens 

 play an important part in the identification of this very variable species. 



Only such variations are here cited as are trees, and the writer is sorry to 

 state that, owing to incomplete material from other varieties occuring in the 

 Kohala Mountains, Hawaii, and West Maui mountains, certain trees are here- 

 with omitted. They are, however, all referable to Tetraplasandra meiandra. 

 It is the writer's intention later to monograph this interesting family. 



On Hawaii on the lava fields of Puuwaawaa, North Kona, grow r a few speci- 



,347 



