16 



by a three-lobed bract, but altogether the morphology of the CiipuH- 

 ferce is rather difficult. 



The keys of the ash may be considered as winged fruits or samarae, 

 the wall of the fruit being extended into a winged expansion whose 

 resistance to the air, as it falls, allows the seed to be carried away 

 from the shade of the parent tree. 



In the elm we get fruits with a more rounded wing, and in the 

 birch also we get little samar^e. 



There are a number of fruits which separate into little portions 

 corresponding to the carpels, but these do not liberate their seeds, 

 and so do not come under our definition of dehiscent fruits. These 

 are called schizocarps. A good example is seen in the hollyhock 

 (PI. VI, fig. 4). In the lower fruit the epicalyx is turned down 

 and part of the calyx removed and also some of the segments, so 

 that the structure can be seen. Each little segment contains a seed, 

 and separates from its neighbour and the central axis. 



In the sycamore (Plate VI, fig. 5) we have a schizocarp which 

 divides into two parts, each of which is sometimes called a samara. 

 The success of the well-known arrangement for distributing the 

 seeds is evident to anyone whose garden is near a sycamore ; the 

 little seedlings are found growing up all over the place. There is a 

 similar structure, in the maple, but the samaras are arranged more in 

 a straight line. 



In the Uiiibellifera the special name of cremocarp has been given 

 to the fruit, which splits into two halves or mericarps. 



In TropcEohim (PI. VI, fig. 6) the fruit separates into three parts. 

 The walls of these are not very hard, so some may, from their 

 definition, include them among the succulent fruits, but they come 

 conveniently in here with the other schizocarps. 



I now pass on to the dehiscent fruits, or those which open to 

 liberate their seeds. There are a large number of these, but they 

 can be grouped fairly easily into a number of varieties. 



The simplest form will be that in which we have a single carpel, 

 which opens down one side, as in the monkshood (PI. VI, fig. 7). 

 Each of the separate parts is called a " follicle." If we regard one as 

 a leaf which is folded inwards from the mid-rib we get the edges 

 uniting at the ventral suture, while the mid-rib is represented by the 

 dorsal suture. The seeds are attached to the edges down the ventral 

 suture, and it is along this line that the follicle dehisces. 



If the carpel opens down both dorsal and ventral sutures, as in the 

 pea-pod (PI. VI, fig. 8), the fruit is called a " legume." Here, also, 

 as in the follicle, it is on the sides of the ventral suture that the peas 

 or seeds are attached. This form of fruit is characteristic of the 

 Legi/minosce, the name of the order being derived from that of the 

 seed-vessel. 



We now come to those dehiscent fruits which are derived from 

 a compound or polycarpellary ovary. 



In the wallflower (PI. VI, fig. 9) there is a two-celled fruit formed 



