1854.] Linnean Society. 335 



placed on such definitions, especially in regard to the terms base and 

 apex' of the seed, because in the figures there given the seed is placed 

 in a position diametrically opposite to that in which it is attached to 

 the placenta, and there is therefore an evident misconception of the 

 whole structure. Von Martins, in his admirable work on the plants 

 of Brazil, gives no account of the structure of the seed of the many 

 Guttiferous genera he there details ; but he describes minutely the 

 seed of Platonia insignis, where the nucleus, enclosed within the 

 testa, is stated to consist of a large mass of fleshy albumen, having 

 in its centre a long terete embryo with a superior radicle, the whole 

 consolidated into one integral inseparable mass. As this form of 

 embryo was opposed to the general conclusion of botanists in regard 

 to the structure of the seed in ClusiacetE, he suggested the propriety 

 of placing the genus Platonia in a distinct family, which he called 

 Canellacea, thus associating it with Canella alba, a plant very diff"erent 

 in habit and floral structure, and of which little is known of its carpo- 

 logical structure. Endlicher, in his ' Genera Plantarum,' gives the 

 character of each genus of the ClusiacecB in accordance with the views 

 of CambessMes, and arranges Platonia, following Martins, in the 

 Canellace<je, as a suborder of Clusiacece. Poppig, in describing several 

 Guttiferous plants of Peru and Brazil, gives no account of the struc- 

 ture of the seed. Lindley, in his 'Vegetable Kingdom' adopts the 

 views of Cambessedes in regard to the Clusiacece, at the same time 

 that he admits Platonia as a member of that family. Miquel, describing 

 in 1844 a species oi Arrudea, in like manner misconceived the struc- 

 ture of the seed, attributing to it an embryo with plano-convex cotyle- 

 dons and a very short radicle. Lastly, Choisy, in a more recent memoir 

 (1850), while he treats at some length on the organography, aflini- 

 ties, and subdivisions of the Guttiferae, and gives differential characters 

 of the several genera of the family, nowhere alludes to the structure 

 of the seed, which is the more remarkable, because the facts pub- 

 lished in the thirty years elapsed since his first memoir are com- 

 pletely at variance with his former views on the subject. 



Mr. Miers then proceeds to give the results of his own observa- 

 tions on the structure of the seed, selecting first that of a species of 

 Clusia, closely allied to Clusia criuva, upon which Cambessedes prin- 

 cipally relied in the construction both of his ordinal and generic 

 characters of the family. The seed-vessel is here described as being 

 formed of five fleshy valves, which break away from a central five- 

 winged persistent column, in the angular recesses of which several 

 seeds are horizontally attached in two longitudinal rows. Each seed 

 is about an eighth of an inch in length, is oval, slightly gibbous on 



