204 Mr. J. Miers on the genus Aptandra. 



hypogynous scales are not exterior to the stamens, but alternate 

 with them, forming one common whorl ; the anthers are introrse 

 and somewhat 4-celled, and their mode of dehiscence, although 

 sometimes valvular, is very different, and finally the leaves are 

 furnished with stipules. 



In Bruniacece we meet mth extrorse stamens, but they offer few 

 other points of analogy ; the ovarium is there inferior, and they 

 have quite a different habit. 



The AlangiacecB present some few points of resemblance, in 

 the form and aestivation of their corolla, in the union of the an- 

 thers into a tube, and in their ovarium with two suspended ovules; 

 but the calyx is wholly adnate with the ovarium, the filaments 

 are free, the introrse anthers burst by longitudinal slits, and the 

 ovarium is distinctly bilocular. 



The OleacecB, especially Chionanfhus, Linociera, and Tessa- 

 randra, offer some degree of similitude, in the form of the calyx 

 and corolla, but their ovarium is bilocular, the ovules are placed 

 collaterally in pairs in each cell, the stamens are few and free, 

 they want the petaloid scales, and finally they have opposite 

 leaves. 



Leonia presents stamens with the filaments united at base, but 

 the tube thus formed is adnate upon a gamopetalous corolla, and 

 the structure of the anthers is wholly different. 



The same objections may be offered to the Styracece, although 

 they have often extrorse stamens. 



The anomalous genus DicUdanthera has its anthers furnished 

 with reflexed valves, M'hich, as in Aptandra, open from the top to 

 the bottom ; but they are introrse, and by the adhesion of the 

 filaments to the petals, appear sessile in the mouth of a gamo- 

 petalous corolla, and it offers otherwise few analogies. 



There are some points of accordance in the Sauvagesiacece, in 

 their internal row of petaloid scales, sometimes combined into a 

 tube, and in having the stamens opposite to the petals. The 

 anthers are extrorse, and even confluent into an incomplete tube 

 in Luxembergia ; there exists also some analogy in their ovarium 

 being 3-celled at base and unilocular at summit, but they differ 

 in their imbricated calyx and corolla, distinct stamens, the pa- 

 rietal placentation of the ovarium, and their remarkably stipulate 

 leaves. Luxembergia however is placed by M. Planchon, with 

 much reason, among the Ochnacece. 



The OlacacecB present many strong points of resemblance, for 

 we have there, as in the genus under consideration, a small calyx 

 with minute teeth, equal in number to the petals, which are ge- 

 nerally four in number, often linear, of thickened texture, and 

 valvate in aestivation ; they have also free appendages of various 

 forms alternating with the stamens ; these last-mentioned organs 



