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



valves and four septa*, and in fact the genus Tetrapoma is of this 

 kind, in which the number of parts which the silicule usually 

 possesses is exactly doubled. Bernhardi therefore in his latest 

 memoir on the subject t regards the perfect ovary of Cruciferae as 

 composed of four pieces with the same number of septa, cells and 

 placentas, but two of which are usually suppressed ; not explain- 

 ing himself definitely as to the origin of the septum, that is, whe- 

 ther it is a prolongation of the valves of the fruit or of the axis. 

 The latter view however seems to be most favoured by Bernhardi, 

 and in fact it is also in my opinion that alone which is sufficient 

 to explain the structure of the fruit of the Cruciferse. In them, 

 says Aug. St. Hilaire, the axis divides into two branches which 

 traverse the ovary and again unite at the point to form the style. 

 These are the two pistillary cords which bear the ovules, and the 

 carpellary leaves, which when ripe separate from them, are inde- 

 pendent of them. The ovary has therefore the simplest possible 

 structure here ; it consists of two carpels and two parietal pla- 

 centas {. According to this view the septum must be regarded 

 as the substance of the axis compressed to a mere plate, in and 

 on which the cellular tissue, conducting the fertilizing matter, is 

 prolonged from the style, and which in fact is itself only a modi- 

 fication thereof for this particular purpose. The double lamellae 

 of the septum can as little be made an objection to this view, as 

 the often remarkable attenuation, slits or opening in the middle 

 of it, or even the total absence of a septum ; since, as to the first, 

 each of the two cavities should be formed independently ; and 

 with regard to the second, it is well known that it does not occur 

 more frequently than cavities in the middle of the pith, which 

 may reach to a total disappearance of the same. 



From this point of view also I hold the anomalous forms which 

 occur in the silique to be most satisfactorily explicable. Assuming 

 as the normal structure, that a silique is two-celled with a per- 

 pendicular septum, it will become one-celled and single-seeded 

 either by the total absence of a septum, the place of which is oc- 

 cupied by the cavity with the single seed, or by only one remaining 

 in the ripe fruit of two one- or more -seeded cavities, since gene- 

 rally only one seed becomes developed. The first case occurs in 

 Isatis, Clypeola, Tauscheriay here there is in reality no trace of 

 a septum, and the single seed hangs down therefore from the 

 apex of the cavity, which usually occupies the centre of the sili- 

 cule. It is the same condition that we meet with in the ano- 

 malous fruits of Althionema heterocarpum, with the difference that 



* Flora 1838, no. 9. 



t Ueb. d. Metamorphose d. Pflanzen ; Flora 1843, Nos. 3, 4. 



