244 



pacity of naturalist. It bears a close resemblance to some species of Disporum ; and 

 it moreover agrees with tliat genus in its sepals being produced into a pouch at their 

 base. The flowers, which are also terminal and in pairs, are twice the size of those of 

 the preceding, and the style is copiously hairy.'' — p. 534. 



From their position in the natural system, and the author's remarks 

 on their affinities, it will be evident that the plants described in the 

 papers above noticed bear a close relationship to many interesting 

 British genera, including Trichonema, Convallaria, Ruscus, Paris, 

 Tofieldia, Narthecium, &c. 



Art. XXXV. — On some new Brazilian Plants allied to the JVatural Order 

 Burmanniacese. By John Miers, Esq., F.L.S. 



Von Martius, in his ' Nova Genera et Species Plantarum Brasi- 

 liensium,' gives the characters of the genus Burmannia, and fully 

 describes five species, discovered by him in the interior provinces of 

 Brazil. Michaux gave the generic name of Tripterella to two North 

 American species, which Mr. Miers appears to consider as not dis- 

 tinct from Burmannia. Seven other species have also been found in 

 Africa, India, and New Holland. Previously to his departure from 

 Brazil the author discovered five new plants, closely allied to Bur- 

 mannia, but differing from that genus in many important particulars. 

 These five species, together with another discovered by Mr. Schom- 

 burgk in British Guinea, and Nuttall's Apteria setacea, are divided 

 into three genera, namely, Dictyostega, containing four species, Cym- 

 bocarpa with one, and Apteria, Nuitall, (formerly Stemoptera, Miers) 

 with two species. All these genera and species are fully described in 

 the present paper, and their characters minutely illustrated by figures. 



The author observes that the Burmanniacese may be divided into 

 two groups. The first will contain the genera Burmannia and Gony- 

 anthes, having a trilocular ovarium and central placentation ; the se- 

 cond, possessing a unilocular ovarium and parietal placentation, will 

 include Dictyostega, Cymbocarpa, Apteria and Gymnosiphon. If the 

 principle be adopted " on which Apostasieae have been separated from 

 Orchideae and Xyridese from Restiacese," the two sections must be kept 

 distinct, and the author suggests that the second would, in that case, 

 form a separate family, under the name of Apteriaceae ; if however the 

 difference in the structure of their ovaria be not thought sufficient to 

 warrant their separation, they must remain associated as Burmannia- 

 aceae, the first section being named Burmannieae, the second Apteriese. 

 The author mentions Gentianeae as an order presenting many similar 



