FRUIT. 



95 



of the Chestnut be examined before or soon after impreg- 

 nation, it >yili be found to contain fourteen ovules in seven 

 cells ; but in the progress of development it becomes one- 

 celled, and thirteen of the ovules are obliterated. Tiic ova- 

 ry of the Oak is originally tluec-celied, willi six ovules; 

 but when perfected it is one-celled and one-seeded. There 

 are many cases of this kind. The reverse of this takes place 

 in some cases, which would be inexplicable, were not the ova- 

 ries examined in their earliest slate. A one-celletl ovary 

 becomes a two or more celled tiuit: in the Crucifertc by the 

 enlargement of the placenta ; in the Astragalus by the ex- 

 pansion of the suture, in other cases by the dilations of the 

 lining of the i)ericarps which form horizontal dissepiments. 

 The Pomegranate presents a remarkable variation from the 

 true type. 



113. When the fruit arrives at maturity the pericarp ei- 

 ther bursts or it remains closed ; if the latter, it is said to be 

 indeh'tsccnt, as in the apple, hazle-nut, &;c. If it bursts it is 

 said to be dchisccntj and it follows invariably the same course 

 in the same species; hence it is important to notice the va- 

 rieties. In some cas<^s the dehiscence takes place by dividing 

 the dissepiments, tliat is tlu^ ( arpels separate into their orig- 

 nal leaves, as in the Delphinum, and this is called sf pticidal 



dehiscence. Fig. 121 represents this kind 

 ^ in which a represents the axis, d the dis- 

 sepiment, and V the valves. In other ca- 

 ses the dissepiments are attached to 

 the middle of the valves, and the dehis- 

 cence in such cases, no doubt, takes place 

 at the midrib of the leaves that i'orm the 

 carpels. This is called loculicidal dehis- 



cence, as in the Lilac, 



Gladiolus &c. fig. 122. 



Septifragal dehiscence 



is when the dissepi- 



ments separate from 



the valves, but adhere^ 



to the axis as in the' 



Convolvulus, fig. 123. 

 In sutural dehiscence there is but one carpel, and of course 

 no true dissepiments, as in the Pea and Bean. Tiie circiim- 

 scissile dehiscence occurs by a transverse separation of the 

 valves half round the pericarp, as in the Anagallis. This is 

 an uncommon mode. The Plantagohas'a/ra«5rer5c dehiscence. 

 Besides the above modes of opening, the pericarp is often 

 rupturcdf produced by a contraction of a portion of it, and 



