352 



BRASSICACEiE. 



[Hypogynois Exogen* 



posterior sepals, the two pairs that remain belonging iii fact to the four petals. But it 

 is in their fruit that the great peculiarity consists. 



Sv,^ 



Fig. CCXLIII. 



Since the placentae are opposite the lobes of the stigma in this 

 Order, it is difficult to reconcile the fruit with any general 

 theory of structure. Either it is in reality composed of four 

 carpels, two of which are abortive, as was first suggested by me 

 in the Botanical jRegister, fol. 1168, or each of the two lobes 

 of the stigma is composed of two half lobes belonging to different 

 carpels, as in Poppyworts. In any view, the dissepiment which 

 cuts oft' the interior of the fruit into two cells must be considered 

 spurious, and a mere expansion of the placentte. 



The opinions of Botanists are much divided as to this matter ; 

 M. Kunth agrees with me in considering the fruit composed of 

 four carpels. And a variety of evidence has gradually collected 

 in favour of this theory. M. Alph. De Candolle has shown that 

 the common Wall-flower is occasionally 4-celled (Monstruosites 

 Vegttaux, 15. t. 5.) There is a genus called Tetracellion, which 

 derives its name from the same circumstance. Mr. Barker 

 Webb has published an account of a Canary shrub, named 

 Parolinia, in which the valves are constantly extended into 

 stigmas. But Mr. Howell (Ami. N. Hist. x. 254.) adopts 

 Brown's view of the subject, and, because of the supposed 

 affinity of Poppyworts, concludes that the fruit of Crucifers is 

 only composed of two carpels. He does not, however, offer any 

 direct proof of the correctness of this opinion. 



Almost all Crucifers are destitute of bracts, and have the 



Pig. CCXLIII. — 1. Cheiranthus cheiri ; 2. its stigma : 3. the same with one valve off ; 4. a cross sec- 

 tion of a seed ; 5. a diagram to illustrate the position of the parts of the flower. 

 Fig. CCXLIV.— Fruits of Parolinia ornate, after H'i66. 



Fig. CCXLIV. 



