PKNTASAOHilE WALL. AND SPILADOOORYS RIDL. 15 



between them. Ovaries, with such a canal at the top, Linnpeus,* 

 Delarbre,f Withering,! and Wycller§ have noticed; these I hope 

 to discuss in a second paper. 



We may sum up the abnormalities as follows : — 



36 flowers abnormal in number of sepals. 



37 ,, ,, ,, ,, petals. 

 35 ,, ,, ,, ,, stamens. 



38 ,, ,, ,, ,, staminodes. 

 479 ,, ,, ,, ,, carpels. 



6 ,, exhibiting abnormal metamorphoses. 



7 ,, ,, ,, forms of cohesion and adhesion. 

 I wish particularly to lay stress upon the great variability of the 



ovary, and will add here that flowers on the same root were observed 

 to possess three and four carpels, or four and five carpels; and that 

 among the few flowers examined above 700 ft. in the Grampians, 

 I find the same free variability — both increase and decrease in the 

 number of parts — as at Scarborough. 



PENTASACHME Wall. AND SPILADOOORYS Ridl. 



By n. SCHLECHTER. 



The genus Pentar,achme (or, as spelt by Bentham in the Genera 

 Plantarwn, vol. ii. p. 769, Peutasacme) was founded by Wallich in 

 1834, and published by Wight in his ( 'ontrihntions to the Botany of 

 India, p. 60, where two species, P. Wallicldi Wight and P. caudata 

 Wall., are described. The former was figured in one of the 

 Avonderfully executed plates of Riocreux in Delessert's Icones 

 Sel.ect(B (v. t. 87). In 1848 Decaisne, when writing his Monograph 

 of the Asclepiads for DC. Prodromus, accepted the genus, and gave 

 descriptions of the two species (p. 627), but unfortunately added 

 two more from Cliina, which, as neither of the type-specimens had 

 flowers, will most likely always remain doubtful. Ten years later 

 Bentham published iu Hooker's Kew Journ. Bot. v. 54, another 

 species, P. Championi, which had been collected by Champion in 

 Hong Kong. The distinctions, however, between his plant and 

 P. caudata were so slight that the learned author himself suggested 

 the possibility of its being only a variety of the latter, and repeated 

 the same suspicion in describing the plant in the P'lora Hong- 

 kougensis. 



* Amoenitates Academicce, vi., Leyden, 1764, p. 301. 



t Flore de la ci-devant Auvergne, 2ncl ed., Riom, 1800, i. p. 293. The 

 account of the pollination given here is interesting, because it ignores so 

 markedly C. K. Sisrengel's more correct account in his Entdeckte Gehehnniss d. 

 Natur, in Bau u in Befruchtung d. Blitmen. 



I Botanical Arrangement of British Plants, 2ud. ed., 1787, Birmingham, 

 i. p. 325. 



§ ' Morphologische Beitriige,' loc. cit. 



