Schumacher ia.] flora indica. 65 



plants, before the rhizoma ia developed, the leaves are considerably smaller, propor- 

 tionally narrower, more rugose, and sometimes bullate. 



3. A. lanceolatum (Hook. Ic. Plant, sub t. 157); foliis anguste 

 lanceolatis acutis siiiuuto-dentatis distanter nevvosis superne glabris 

 nitidis snbtus ad nervos adpres&e pilosis. 



Hab. In Zeylaniie inontibus teinpei'atis, W'ujhl ! TJmaitea! — {v. s.) 



Folia anguste spathnlata vel fere linearia, 3-6 poll, louga, i poll, lata, superne 

 argute sinnato-dentata, dentibus glandula apiculatis basin versus" angustata dentibus 

 obtusioribus;/)e/iV)& abbrevialis, alatis. Injlorescentia A.unijtori. Pedicelli 1-2- 

 poUicares, laxe patentim pilosi. 



This species is only known from a few very imperfect specimens in the Hookerian 

 Herbarium : these appear to be young plants, the rhizoma being scarcely deve- 

 loped. There are traces of an inflorescence like that of the last species, but no 

 flowers in a state fit for examination. 



§ 3. Folliculis 1-2-sperinis. 



4. A. costatum (Jack, Mai. Misc., et in Hook. Bot. Misc. ii. 82) ; 

 foliis obovatis basi sagittatis, vacemis scapiformibus erectis laxis, bnic- 

 teis lanceolatis non inibricatis, Horibus breviter pedioellatis. — // a/^. Cat 

 1117 A ! A. 'Wightianum, 7r. et A. Prod. i. 6 ; JJ'ight! III. i. 9. 



Hab. InTravancor, Wight!; Malaga, ad Penang et Singapiir, Jack! 

 Wall.!— {v. s.) 



Rtdzoma lignosum, subhorizontale. Folia 4-8 pollices longa, J-3 lata, dentato- 

 scrrata, scabra, superne sccus costam inter nervos el versus margines raolhter pilosa, 

 subtus pallida, secus nervos adpresse-pilosa. PetioK brevissimi, auriculati, vagi- 

 nantes. Scapi (cum pedicellis et caiycibus) patentim pilosi, supra medium floriferi ; 

 pedicelli bractcis duplo loiigiores. hepala -l^-polHcaria. Stamina 15. Ovaria bi- 

 ovulata, ovulis axi iusertis coUateralibus adscendentibus. 



The materials at the disposal of Wight and Arnott at the time of the publication 

 of the Prodronms were so imperfect that they did not discover that their specimens 

 belonged to two different species. One specimen (belonging to the present species), 

 which had good flowers and fruit, was employed for the analysis of the flowers given 

 in the Prodromus, but all the others belonged to A. ArnoUianum, which alone occurs 

 in the AVallichian Herbarium, under the name of A. Wighiianum. Dr. AYight had, 

 however, retained in liis onn collection the specimen of A. cosialum, along with one 

 of those of A. ArnoUiann-in; and when he had occasion to revert to the subject for 

 the ' Illustrations,' having acquired additional materials, he detected the differences, 

 which he has clearly indicated in that work. Dr. "Wight has also pointed out the 

 probable identify of the A. JFightianum of the Prodromus with A. coslalum. Jack ; 

 and after a comparison of the solitary specimen from Travaneor in the Wightian 

 Herbarium, with those of .Tack and Wallich, we can find no diiferences. As the de- 

 scription of A. V'ic/litianum in \V. A. Prod., which must be considered the authority 

 for the species, agrees in all essential points with A. costatmn, the former name must 

 necessarily be suppressed. 



4.. SCMUMACHERIA, Vahl, Aruott. 



Sepala 5. Petala 5. Stamina indefinita, unilateralia, monadelpha, 

 pluriserialia, filamentis in columnam brevem oblique cylindricam coalitis. 

 Antlierm subsessiles, lineari-oblongas, obtusse, apiculatae, biloculares ; 

 loculis lateraliter dehiscentibus. Ovaria 3, discreta, dense pilosa, uni- 



K 



