II.] 



CLEISTOGAMOUS FLOWERS. 



33: 



belonging to very different groups. They were, I 

 believe, first observed by Dillenius in Ruellia {Hortus 

 Eltlianiensis, v. ii. p. 239). As, however, the}- have no 



Fig. 36. — Cleistogamous flower of 

 LaiiLiiaii aiiiplexicaute. 



Fig. 37. — S«ctiun o) ditto. 



relation to our present subject, I shall not now dwell 

 upon them. 



I pass on, therefore, to the genus Primula. If a 

 numljer of specimens of primroses or of cowslips be 



OoOO 

 ° 



X zso 



Fig. 38.— Primula (long-styled fortn). FiG. 39.— Primula (»hort-styled form). 



examined, we shall find that about half of them 

 have the pistil at the top of the tube, and the 

 stamens half-way down (as is shown in Fig. 38), 



D 



