218 



MANUAL OF BOTANY 



the allied orders, the Valerianaoese and Dipsacese, it is sus- 

 pended or pendulous. In the Polygonacese (fig. 476), the ovule, 

 which springs from the axis, is also always solitary and erect ; 

 and in the Thymelaceae {fig. 479) it is suspended. In other 

 natural orders we find the position varying in different genera, 

 although generally constant in the same genus ; thus, in the 

 Eosaces, the genera Oeuni, Alchemilla, and others, have an 

 ascending, while those of Poterium, Sanguisorba, &c., have 

 a suspended ovule, and in Potentilla both are found. In the 

 Ranunculacese also we find the ovule varying in like manner 

 as regards its position. 



Fig. 477. 



Fig. 478. 



Fig. 479. 



Fig. 480. 



Fig. 477. Vertical section of the ovary of the Mare's Tail {Hippuru vul- 

 garis). 0. Ovule, which is inverse or pendulous, and anatropous. .?. Base 



of the style. /. Funiculus, r. Haphe. c. Chalaza. Fig. 478. Vertical 



section of the ovary of the Pellitory (Parietaria officinalis), with a single 

 ascending ovule. The letters have the same references as in the last figure. 

 Fig. 479. Vertical section of the ovary of the Mezereon {Daphne Meze- 

 reum). containing a solitary suspended ovule. The letters refer as before. 

 From Jussieu. Fig. 480. Vertical section of a cell of the ovary of a spe- 

 cies of .^senilis containing two ovules, o, o, one of which is ascending and 

 the other suspended, m, m. The micropyle or foramen in the two ovules. 

 s. Base of the style. From Jussieu. 



When the ovary or cell has two ovules, these may be either 

 placed side by side at the same level and have the same 

 direction, as in Nuttallia, when they are said to be collateral ; 

 or they may be placed at different heights, and then they may 

 either follow the same direction, when they are superposed ; or 

 one ovule may be ascending and the other suspended, as in 

 Mscvlus (fig. 480). The position of the ovules in those cases 

 where they are in definite numbers, is also usually constant 

 and regular, and similar terms are employed ; but when the 

 number of ovules in the ovary or cell is indefinite, the relations 

 are less constant, and depend in a great measure upon the 

 shape of the cell and the size of the placentas. Thus in the 



