364 M. Treviranus on the Structure of the Fruit of the Cruciferse. 



and parenchyma, as these agree in their general peculiarities with 

 those which are so called in leaves. The epidermis forms two 

 lamellae, whether the septum be thick or thin, between which 

 the parenchyma, which however is often at least partly wanting, 

 is inclosed and distributed in different degrees and abundance. 

 The cells of the first-named substance differ much in their form, 

 position and mode of union. Very often their circumference is 

 more or less drawn out lengthways, as in Cheiranthus Cheiri, 

 Lunaria annua, rediviva, Farsetia clypeata, Vesicaria utricalata, 

 sinuata, &c, and the longer diameter is then never parallel with 

 the axis of the silique, but at a sharp angle to it. The direction 

 however is different in the two lamellae, and therefore the reticu- 

 lations, which from the transparency of the septum are visible at 

 the same time, never correspond, but cross one another, a cir- 

 cumstance which Brown* has remarked, and which appears to 

 favour the view taken by DeCandollef, that the two lamellae be- 

 long to different carpels. In the very thin septa of Draba, Cap- 

 sella, JEthionema and Camelina, the borders of the cells resemble 

 those of the epidermis of delicate leaves, being undulated, which 

 Brown J calls amorphous ; in those of a firmer structure, on the 

 contrary, for instance in Cheiranthus and Lunaria, they have a 

 straight course, and are then more or less knotty. This knotty 

 structure, which is not observable in the earliest stateof the septum, 

 but is formed subsequently, depends upon a thickening of the 

 cell-walls with a simultaneous perforation of them by canals, 

 which run from within outward in a manner resembling that 

 which Meyen delineates as the structure of the punctated cells of 

 some plants § ; but usually the pore-like marks, lying in long 

 rows, are only observable when the wall of a cell is seen from the 

 side, where it is united with another. Several observers have also 

 remarked pores upon the septum usually similar to the pores of 

 epidermis, for example Hartig and Schleiden in Capsella Bursa- 

 pastoris, Trecul in Cheiranthus Cheiri ; I also have perceived the 

 same in the last-named plant, and most clearly in Octadenia ly- 

 bica, Br. ; in most of the Cruciferse however I have sought them 

 in vain, and hitherto have only observed them in the vicinity of 

 the border of the septum, never in the middle of it. Here in the 

 middle the two lamellae lie pretty close together, while at the 

 borders they inclose, in common with the placental nerves of each 

 side, a space filled with a parenchyma which from thence spreads 

 out further into the septum. This has some affinity to the par- 

 enchyma of the under side of a leaf, consisting of anastomosing 

 rows of elongated cells, which sometimes even contain granular 



* Observ. on the PI. of Central Africa, 13. 



+ L. c. 190. I L. c. 



§ Physiol, i. t. 1. f. 2, 5, 10, 11. 



