GRAMINEAE 



97 



BAM B USA Schreber 



BAMBUSA SPINOSA Roxb. Hort. Beng. (1814) 25 (type!). 



Bambusa spinosa Blume ex Nees in Flora 8 (1825) 580. 



Bambusa blumeana Schultes f. Syst. 7 2 (1830) 1343. 



Bambusa teba Miq. Fl. Ind. Bat. 3 (1857) 418 (type!). 



Arundarbor spinosa Rumph. Herb. Amb. 4: 14, t. 3. 

 Amboina, back of the town of Amboina, Robinson PI. Rumph. Amb. 608, 

 July 17, 1913, locally known as bambu duri. 



There is no question that this is Arundarbor spinosa of 

 Rumphius, and that it is identical with the widely distributed 

 Malayan Bambusa blumeana Schultes f. However, Bambusa 

 spinosa Roxb. was based wholly on Rumphius in the original 

 place of publication, Hortus Bengalensis (1814) 25, by citation 

 of "H. A. 4. t. 3;" that is, Herbarium Amboinense Vol. IV, t. 3. 

 Robinson,* however, states (p. 418) : "Bambusa spinosa Roxb., 

 as typified by Herb. Amb. 4 : pi. 3, is probably not B. arundinacea 

 Willd., and there are various points in Rumphius's description, 

 which oppose its identification as B. Blumeana Schultes f." 

 Munro t states under Bambusa blumeana Schultes f . : "Blume 

 states that this plant is spinous; and there are some traces 

 of spines on the specimens I have seen; but there is nothing 

 to indicate that it is as spinous as B. agrestis of Poir. and 

 Arundarbor spinosa of Rumph. are described to be." From 

 an examination of the Amboina specimen, Rumphius's descrip- 

 tion, and with a definite knowledge of this plant as it grows 

 in the field, for it is one of the commonest bamboos in the 

 Philippines, I am thoroughly convinced that Arundarbor spinosa 

 is identical with Bambusa blumeana Schultes f., but under our 

 rules of nomenclature, the oldest valid specific name is Bambusa 

 spinosa Roxb. The plant is remarkable for the dense thicket 

 of interlaced, stiff, very spiny branches that surrounds the 

 lower parts of the culms, and these are well represented by 

 Rumphius's figure. Ordinary herbarium specimens, taken from 

 the upper leafy branches often do not present spines, or at 

 least only few and greatly reduced ones, which accounts for 

 Munro's statement, as he saw only herbarium specimens and 

 did not know the plant in the field. 



By Loureiro Arundarbor spinosa was reduced to his Arundo 

 agrestis, Fl. Cochinch. (1790) 57, but Arundo agrestis Lour, was 

 based on specimens from Cochin-China, and apparently is Bam- 



* Robinson, C. B. Roxburgh's Hortus Bengalensis. Philip. Journ. Sci. 

 7 (1912) Bot. 410-419. 



f Trans. Linn. Soc. 26 (1870) 102. 



144971 7 



