MAGNOLIACE^. 159 



with scaly buds, and alternate, petiolate, crenulate, persistent leaves. 

 The flowers, also proceeding from scaly buds, and almost always her- 

 maphrodite, are arranged in bunches as in Euptelea, and appear 

 at the commencement of the vegetation of the season. The scaly 

 bracts, which at first protected them, fall off towards the time for 

 their expansion. We can see from the foregoing that this group, 

 to which the name of Trochodendrece' has been given, includes two 

 quite degenerate genera of Magnoliacece, with diclinous achlamydeous 

 flowers. The insertion of the stamens in IVochodendron, and the 

 concave form of the receptacle, which takes away all appearance 

 of independence from the basilar portion of the carpels,' might 

 authorize our establishing for these a small group of perigynous 

 Magnoliacece. 



V. CANELLA SEEIES. 



The Canellas^ (figs. 211-215) are plants with regular hermaphro- 

 dite flowers. On the slightly convex receptacle are successively 

 inserted a calyx and corolla of free pieces, an androceum and a gynse- 

 ceum whose elements cohere by their edges. The calyx consists of 

 three free persistent sepals,^ imbricated in the bud (fig. 213). The 



Hooker bad first referred Trochodendron as an it gives insertion to organs of such complexity as 



abnormal genus. It appears, however, that a the pistil, and that for a good way up, and even 



second species of this genus has recently been near its edges, it becomes difficult to admit that 



discovered in Japan, T. longifolium Maxiii., it is of appendicular origin. If we suppose the 



known to us only by the mention made of it by organic apex — that is, the deepest point — of the 



MiQUEL, in his work on the Origines de la Flore receptacle of Trochodendroii to be pulled upwards 



de Japon (see Adansonia, viii. 211). and raised a little above the insertion of the 



* Hook. F., loc. cit. stamens, we get a convex receptacle like that of 



' The sort of shallow sac or cup formed by the Illicium or Drimys. 

 dilated receptacle is here, in our opinion, of axial •' P. Bhowne, Jamaic. (1756), 275, t. 27. — 



nature, and its organic base corresponds with the Swaetz, in Linn. Trans., i. (1791), 96, t. 8. — 



level of the insertion of the androceum. Con- Mitke., Syst. Veg., 443. — G.eetnee, Fruct., i. 



sequently this sac is not of foliar origin, and 373, t. 77. — A. L. Juss., in Mem. Miis., iii. 347.— 



hence gives insertion to the carpels. These are DC, Prodr., i. 563. — Endl., Gen., n. 5457. — 



truly free as those of most MagnoUacecB, but the A. Rich., Fl. Cub., 245. — Miees, in Ann. Nat. 



base by which they are inserted is much extended Hist., ser. 3, i. 348; Contrib., i. 112, t. 23. — 



and very oblique. Thus we have a great resem- Payee, Fam. Nat., 102. — B. H., Gen., 121, n. 



blance between the organization of the flower of 1. — H. Bn., Adansonia, vii. 12,67. — Winterania 



Trochodendroii and that of certain Eosacece, L., Gen., n. 598. — Juss., Gen., 263. 

 of Eupomatia among the Anonacece, and of most * To Bextham & Hookee these three leaves 



of the Monimiacece. For it is true that strictly, represent bracteolae, forming a sort of calyculus 



one may consider the sac surrounding the flower under the flower, which would then be apetalous 



in the last named order as a calyx when it only in Canella ; for these botanists call that coloured 



supports stamens ; but when, in the female flower, perianth a calyx which most other authors call a 



