114 Mr. J. Miers on some genera of the Icacinacese. 



very different in habit from any of the Artocarpacea, having ei- \ 

 stipulate leaves, flowers with a regular and symmetrical calyx! 

 and corolla, stamens equal in number to the petals, an ovarium 

 with two anatropal ovules, suspended from the summit of the 

 cell, and a seed with a considerable quantity of albumen, en- 

 closing an embryo with large foliaceous cotyledons, and a small 

 inferior radicle. In regard to the structure of the seed of Phyto- 

 crene, our evidence is yet quite uncertain. Prof. Lindley (Yeget. 

 Kingd. p. 274) describes and figures an albumen of a very gra- 

 nular, or rather ruminated texture, enclosing two large foliaceous 

 cotyledons, with a very small inferior radicle ; and Mr. Brown, in 

 his generic character, greatly confirms this view, by stating it 

 to possess an embryo with large foliaceous cotyledons, enclosed 

 in albumen. Prof. Blume, on the contrary (Mus. Bot. Lugd. 

 Bat. p. 41. tab. 7), describes and figures the embryo as being 

 quite exalbuminous, with large foliaceous crumpled cotyledons 

 of a rugosely granular texture, possessing a short superior ra- 

 dicle : he here acknowledges Phytocrene to be identical with his 

 Gynocephala, the fruit of which he describes as consisting of an 

 aggregation of several elongated drupes, upon a fleshy receptacle, 

 forming a globe as big as a man's head. It must at the same 

 time be acknowledged, that the extremely villous habit of Phy- 

 tocrene, the peculiar structure of its woody stem, its closely ag- 

 gregated flowers in globular heads, the membranaceous texture 

 of its calyx and corolla, both clothed externally with very dense 

 long hairs, and its peculiar stamens, present characters to which 

 little resemblance can be traced in Sarcostigma. I urge these 

 reflections, however, with extreme hesitation, in deference to the 

 conclusions of an authority, whose determinations all botanists 

 will regard with the highest consideration. It is to be regretted, 

 however, that Mr. Brown has not favoured us with his views, 

 and the reasons on which they are based, in regard to the real 

 atfinities of the Phytocrenece ; but he says decidedly that Sarco- 

 stigma, which in his opinion " so obviously belongs to " this 

 group, bears no relation to HeimandiacecB, to which family that 

 genus had originally been referred by Drs. Wight and Arnott. 

 We may, however, infer something more tangible on this point 

 from his admission of "its near relationship" to Pyrenacantha, a 

 genus with a single floral envelope, and other characters, that 

 have led to its position near the Antidesmece. The genera Phy- 

 tocrene, Nansiatum, and lodes form a very natural group, pos- 

 sessed of consimilar features, offering constantly a regular calyx 

 and corolla, divided into segments equal in number to the sta- 

 mens, all alternating with each other in distinct series; they 

 have therefore every claim to rank among the Dialypetalce m 

 Endlicher ; but this disposition does not exist in Miquelia, a ge- 



