17 



from a one-celled ovary by the growth of a false septum across it. 

 The false septum is frequently referred to as the " replum," but this 

 term only properly refers to the framework across which the septum 

 stretches. Dehiscence in the Cruciferce commences at the bottom, 

 extending upwards, but in other dehiscent fruits it generally com- 

 mences at the apex, extending to the base. The long-shaped fruit 

 of the wallflower, etc., is called a "siliqua." 



When the fruit is about as broad as long it is called a " silicula." 

 A good example is seen in honesty (PI. VI, fig. lo). In the upper 

 fruit dehiscence has only just commenced. In the middle one the 

 valves are just adhering at the apex, while in the lowest they have 

 fallen off, leaving the replum with the septum stretched across it. 

 The seeds often remain closely applied to the septum, but here have 

 fallen away. 



In the radish (PI. VI, fig. ii) there is a structure formed as in the 

 other Crucifera;, but it does not open in the same way. It divides 

 off transversely, and is called a lomentaceous siliqua, a lomentum 

 being a variety of the legume (found in acacia), which contracts in 

 a similar manner. 



In the crane's-bill (PI. VI, fig. 12) the carpels separate from one 

 another, as in the schizocarps, but each carpel also opens, and the 

 seed is ejected with some force by the contraction of the walls of 

 the carpel. The carpels remain attached by the upper part of the 

 withered styles. 



There are a number of dehiscent fruits called " capsules," and 

 they are classified according to their mode of dehiscence. 



In the pink family we have a partial dehiscence by means of teeth 

 (PI. VII, fig. 13). Fruits of the campions, if found in wet weather 

 in the winter, will be seen to have these teeth meeting together, but 

 as they dry they open out again. 



A number of capsules with complete dehiscence split open 

 longitudinally. This splitting may occur in three ways — either 

 down the middle of each cell, or at the line of juncture of the several 

 cells, or the outer walls may separate, leaving the seeds attached to 

 a central axis, or to walls radiating from it. 



In the tulip (PI. VII, fig. 14) there are three cells, one of which 

 is seen with the dehiscence down the middle, which represents the 

 dorsal suture. 



The box (PI. VII, fig. 15) is another example of this type. 

 Shortly after bringing some of the specimens home I heard the 

 seeds, which are smooth and wedge-shaped, dropping all about the 

 room as they were shot out to some distance by the contraction of 

 the walls of the fruits. 



The fruit of the horse-diestnut is a further example of the same 

 type. A seed can be seen (PI. VII, fig. 16) with the large hilum 

 or area by which it was attached to the placenta. 



In the violet (PI. VII, fig. 17) we have a one-celled fruit formed 

 from three carpels, and it opens along the lines of the dorsal sutures. 



