C 68 ] 



resemblance vanishes. Hassharl published his species in a 

 supplement to "Flora" (IT. 18&2. p. 5) as follows: 



"Agave Rumphii mihi. Acaulis foliis lineari-lanceolatis, 

 canaliculatis acuminatis spinis marginalibas nigris rigidis sursum 

 areuatis, scapo ramoso d<dn viviparo, staminibus corollam longe 

 super antib us. Corolla? limbo revoluto. Ch. Sprg, S. V. II. 79 

 Willd. Sp. PL II. 195. Scklt. VII. 127. Rumph. Amb. i\ 273. t, 94 

 Norn. Sund. Nannas sabrang = Bromelia transmarina" 



He does not cite A. Cantula Roxb. as a synonym, and the 

 revolute segments of the perianth of his description do not suit 

 any species as yet observed in India. Zollinger (Syst, Vcrz. der 

 in Ind., Archipel, etc., Zurich, 185) sets aside Hassharl' s name 

 in favour of Roxburgh's ; he had at least one specimen before him 

 and had seen the plant of Hassharl living in the Buitenzorg 

 Garden. In the latest Catalogue of that garden (Teysmann and 

 Binmndijk, Cat. PL H. B liogor, 1866) A. Cantula Roxb., is 

 maintained and A. Rumphii, Hassharl excluded. 



In the Garden of Sir Thomas Hanbury at Mentone (Sysi. 

 Cat. by G. Cronemeyer, Erfurt, 1889) Agave Rumphii, A. 

 Cantula, A. vivipara, and A. Theometel are all separately 

 represented. Mr. Watson of the Kew Garden visited the 

 Riviera during 1899 (Kew Bulletin No. 36, Dec. 1889, CXXVI) 

 and his notes on Agaves include the following. 



"A. Rumphii Hassk. Mr. Baker refer* this to *A. vivipara* 

 but the plant under the former name in Mr. Hanbury's Garden 

 looks like a gigantic A. rigida ; it is a very fine Agave, the leaves 

 about 5 feet long and very numerous." 



***** 



"A. vivipara Linn. (A, Cantula). A distinct species, more 

 resembling a furcraea. This lately flowered at Mentone" 



In November 1891 Mr. Baker himself visited the Eiviera 

 (Kew Bull No. 61, January 1892, CCXXII1). 



We extract the following as it is material : 



" A. rigida Miller. This, the most valuable and most variable 

 of all the Agaves, is common and quite at home in the Riviera 

 Gardens, flowering freely ; and I had an opportunity of studying its 

 characters and range of variation far better than 1 had ever done 

 before and of seeing several forms with which 1 was not previously 

 acqauinted. The commonest forms in the Riviera show the charaC" 

 teristic small distant nearly black teeth t and agree very well with 



