M. Treviranus on the Structure of the Fruit of the Cruciferae. 363 



Cardamine, though they are so small here that they escaped the 

 sharp-sighted eyes of Brown and DeCandolle *. They are most 

 distinct in the boat- shaped valves, for example in those of 

 Thlaspi arvense and still more in those of Megacarpcea, where 

 each describes a circle in its course and marks the origin of 

 the broad wing, into which numerous branches are given off 

 from it. In flat or slightly elevated valves also they generally 

 give off many branches at right angles which anastomose with 

 those also coming out at right angles from the placental trunks ; 

 in its further course however the main trunk becomes more and 

 more attenuated, till finally it terminates in a very minute pro- 

 cess inside the apex of the valve of each side without reaching or 

 going beyond the apex itself, which is most striking when this 

 forms a tooth-like projection, often of considerable length, as in 

 the genera Notoceras and Paroliniaf. 



The genus Camelina, however, as Koch rightly observes % t 

 forms an exception to this structure. Here the style is a pro- 

 longation of the two valves of the silique, the vessels of which 

 pass into it, so that on the separation of the valves it splits into 

 two halves, each half having in it the end of one trunk of val- 

 vular vessels. But in Cardamine this trunk is always slender, 

 without distinctly branching, and it is quite lost at two-thirds of 

 the height of the valve. 



This course of the two kinds of vessels of the fruit compared 

 with their mode of distribution in other families, as in Primu- 

 lacese, Gentianese, Scrophulariacese, Saxifragese, Caryophylle8e,&c, 

 differs from them in so far, that in the last-mentioned families 

 the pericarpial vessels alone are continued from the ovary into 

 the style, or pass through the style as far as the base of the 

 stigma, while the placental vessels, which in the Cruciferae exclu- 

 sively supply the style or the stigma with vessels, do not take the 

 least part in it. 



As to the septum, leaving out of the question the funiculi 

 which mostly cohere with it, it has no vessels of its own in the 

 silique ; at least it does not possess them in the sense in which 

 this expression is usually taken, as signifying fibrous and spiral 

 tubes, but has a structure wholly cellular, and this of a peculiar 

 kind. It consists of two substances which I will name epidermis 



* L. c 201. 



f " Mr. Webb has published an account of a Canary shrub named Paro 

 linia, in which the valves are constantly extended into stigmas " (J. Lindley, 

 Veget. Kingdom, 352). Anything but that ; Webb's descriptions and illus- 

 trations rather show in the most distinct manner that these processes of the 

 valves are not stigmata, but unusually large horn-like appendages, in which 

 the mid-nerve of the valve is continued to the very apex, which is split. 

 (Ann. d. Sc. Nat. 2 ser. xiii. 136. t. 3.) 



t Deutschl. Flora, iv. 570. 



26* 



