DICOTYLEDONS. 297 



perlcince, an idea which its structure very much confirms, notwith- 

 standing its difterence in habit. The fringed bodies, which are con- 

 veniently enough called Nectaries, appear so very analogous to the 

 bundles of stamens in Hypericum, that we confess we should have had 

 little doubt in our own minds of its actually belonging to HypericuKe, 

 if the most learned botanist of his age had not formed an opinion to 

 the contrary. 



2. f^MPETRUM. 

 Dicecious. Ca/. 3-parted, persistent. <?• Pe/a/^ 3, wither- 

 ing. Stam^nsS. Filaments long. Anthers 2-part\te. ?. Ova- 

 ry superior depressed. Style none, or very short. Stig- 

 mas nine, reflexed, spreading. Jass. Berry round, l-celled. 

 Seeds 2 — 3 or more, erect, bony. Embryo upright, in the 

 axis of a fleshy albumen. Radicle miexiov . Gcerln. 

 1. E. ?iicrrum, procumbent, leaves linear-oblong, p. 287. — 

 Mountainous heaths. 



Jussieu has placed this singular genus, along with Hudsonia, at the 

 end of Ericece. Adanson arranged it with his Cisti. But we think 

 Mr. Nuttall has taken a more correct view of its affinity, m placing it, 

 as a distinct order which he proposes to call Empetrea, at the end of 

 ComfercE. We quite agree with him in thinking that the principal re- 

 semblance between it and Ericece consists in the leaves, 



3. CERATOPHYLLUM, 

 Monoecious. Cf/Z. many-parted. Cor. 0. S- St am. twice as 

 many as the divisions of the calyx. Anthers oblong, ? . Ova- 

 ry 1, compressed. Sty Iff none. Stigma oblique. Juss. 

 Nut bonv, 1 -celled, indehiscent. Seed erect. Albumen 

 none. Embryo vv'ith 4 cotyledons^ of which two are larger 

 than the rest. Plumula of many leaves. 

 1. C. demersum, fruit armed with 3 spines, p. 2/2. — Ditches. 

 We suspect there is something still to discover, with respect to the 

 male flowers at least, in this genus. Its affinity has scarcely been 

 mentioned. For some time it floated among J»ss?eM'5 Naiades; till it 

 was ascertained that it was a dicotyledonous plant. Since that time, 

 the only author we are acquainted with who has noticed it is M. Ri- 

 chard, who has hinted at a resemblance existing between it and Coni- 

 ferce. 



