344 FLORA INDICA. [NymphceacetB . 



we have been quite nnable to distinguish them in India, or in our stoves, the dif- 

 ferences between them being of degree only, except the colour of versicolor. The 

 carpels vary in number from eight to twenty and even thirty, and the length to 

 which the apices of the stigmatic rays are extended is also extremely variable ; they 

 are sometimes merely blunt points, and in other cases produced into long incurved 

 points : the latter arc the appendiculate stigmata of Roxburgh's versicolor, and, as 

 Plauchon rightly supposes, are very different organs from the true stigmalic appen- 

 dices of N. lotus. N. Hookeriana of Lehmann we collected at Chittagong, and 

 again at the mouth of the Megna ; its flowers varied from rose-coloured to pale purple 

 and light blue, and it entirely accords with Koxbuj'gh's N. versicolor. 



Edgeworth's N. punctata is founded on the erroneous idea that the leaf of iV. stellata 

 is not punctate, which it almost invariably is in all its varieties, though described as im- 

 punctate by De Candolle. One of Edgeworth's three flowers (in Herb. Hook.) is of the 

 variety versicolor, the two others of N. stellata, — a fair proof in itself of these being 

 but one species. Planchon, whose views of the affinities of the species are always 

 correct, has already suggested its being iV. versicolor. In all the varieties the leaves 

 vary from being quite entii-e to toothed along their whole circumference; all the 

 varieties agree in the arrangement of the air-canals in the peduncles and petioles. 



4. N. pygmsea (Ait. Hort. Kew. ed. alt. iii. 293) ; minima, foliis 

 oblougo-orbiculatis integerrimis lobis acutis, staminibus inappendicu- 

 latis, stigmatibus 4-8 late ovatis coohleariformibus. — Bot. Mag. ]525 ; 

 BO. Syst. ii. 58; Froi. i. 116; Led. M. Ross. i. 84. 



Hab. Assam, Jenkins! montibus Khasia, ad Nonkrem in paludibus, 

 alt. 5600 ped. !— (Fl. Aug.) (u. v.) 



DisTEiB. Sibiria ! China borealis ! 



Rhizoma snbperpcndiculare, dianietr. poUicis, pilis atris moEibus lanatum. Petioli 

 graciles. Folia li-2 poll, longa, elli))tico- v. obovato-orbiculata, lobis divergeutibus 

 acutis, nervis filiformibus. Flares albi, inodori (valde odori, fid. DC), ly- 3 poll, 

 diametro. Calyx basi quadrafcus ; sepalis lineari-oblougis obtusis. Vetala sub-10, 

 sepalis paullo longiora v. iis requilonga, lineari-oblonga, obtusa. Stamina 3-4-seriata, 

 brevia, antheris connectivo a:quilatis, filainentis late dilatatis intimis ad apicem 

 ovarii insertis ; polline suhgranuloso. Stigmatis radii breves, obtusi. 



This curious and well-marked little species is one of the many proofs of the inti- 

 mate relation between the Khasian and Chinese Floras, to which we have alluded at 

 p. 105 of our Introductory Essay ; w^e are unable to find any character by which to 

 distinguish this plant from the Siberian and Chinese, except the inodorous flowers, 

 which tends to weaken that analogous mark of difference between the N. cierulea of 

 Egypt and iV. stellata of India, and the N. alba of Europe and N. odorata of North 

 America. 



2. EURYALE, Salisb. 



Sepala 4, margini tori ultra ovarium producti inserta, ereota. Peiala 

 indefinita, sepalis breviora, 3-5-seriata. Stamina indefinita, multiseriata, 

 seriebus 8-meris, filamentis linearibus; pollen sphsericum, 3-nucleatum. 

 Ooarium 8-loculare, toro apice dilatato immersum ; stigmate discoideo 

 obscure globoso depresse concavo, tubo tori accrete. Oviila pauca, pa- 

 rietibus affixa. Bacca spongiosa, irregulariter rapta, sepalis persistenti- 

 bus coronata. Semiiia 8-20, ariUo pulposo involuta; testa atra orassa. 

 — Herba aculeis liorrida, rhizomate crasso fihras crassas emittente, foliis 

 orhiciilarihm primum corruffatis demum hidlatis marginihus planis, floribus 

 pitrpureo-violaceis suaveohntibus, seminibus edulihus. 



A very remarkable plant, closely allied to the Victoria of the South American 



