CANNACEAE 



165 



pelago, has flowers four times as large as the Javan and Amboina 

 forms, which again differ from each other. Besides Costus 

 speciosus I think there are two other known species common in 

 the Malayan region. One is Costus sericeus Blume (argyro- 

 phyllus Ridl.), the other the form represented by Rumphius's 

 figure, which will probably need a distinctive specific name. 



TAPEINOCHILUS Miquel 



TAPEINOCHILUS AN AN ASS A E (Hassk.) K. Schum. in Engl. Bot. Jahrb. 

 27 (1899) 349; Engl. Pflanzenreich 20 (1904) 436. 



Costus ? ananassae Hassk. in Abhandl. Naturf. Gesellsch. Halle 9 



(1866) 335 (Neue Schliissel 191) (type!). 

 Tapeinochilus pungens Miq. in Ann. Mus. Bot. Lugd. Bat. 4 (1868-69) 



101, t. 4. 



Tubu tubu Rumph. Herb. Amb. 7: 52, t. 22, f. 2. 



The original description of the species proposed by Hasskarl 

 was based entirely on Rumphius. The species was not published 

 by Hasskarl under Tapeinochilus as indicated by K. Schumann, 

 but under Costus. Neither of the names appears in Index Kew- 

 ensis of in its supplements to date. 



CANNACEAE 



CANNA Linnaeus 



CANNA INDICA Linn. Sp. PL (1753) 1. 



Cannacorus Rumph. Herb. Amb. 5: 177, t. 71, f. 2. 



Amboina, Roemah tiga and about the town of Amboina, Robinson PI. 

 Rumph. Amb. 410, August 15, 1913, along river banks, in sago swamps, etc., 

 locally known as tasibe mera. 



This is the common, spontaneous, Indo-Malayan form with 

 small red flowers that is generally placed under Carina indica 

 Linn. As noted by Kranzlin, in Engl. Pflanzenreich 56 (1912) 

 59, it is very difficult to determine the exact status of Canna 

 indica, as described by Linnaeus. The Linnean species, as 

 here interpreted, is apparently not the form described by 

 Kranzlin. Cannacorus was originally reduced by Linnaeus to 

 Canna angustifolia Linn., in Stickman Herb. Amb. (1754) 20, 

 Amoen. Acad. 4 (1759) 130, Syst. ed. 10 (1759) 842, which is 

 certainly an error; in the second edition of his Species Planta- 

 rum (1762) 1, however, he placed Cannacorus under Canna 

 indica Linn., which seems to be the correct disposition of it. 

 By other authors it has been placed under Canna patens Rose, 

 C. orientalis Rose, and C. coccinea Ait. 



