io8 Flowers and their Unbidden Guests. 



it. The base of the ovary therefore which they thus 

 grasp completely occludes the two nectar-pits. Here 

 again an insect, if it would get at the nectar, must force 

 the two appendages from their contact with the ovary. 



Another and similar case is that of Ophelia Wilfordii 

 Kern. In this flower the nectar is secreted in an 

 appendage, situated above the base of the petal and 

 hollowed out into a kind of groove. The anterior 

 and narrower part of this groove is closed in by its 

 margins being elevated into ridges which converge 

 above it, these margins being moreover set with short 

 interdigitating fringes ; the posterior part of the nec- 

 tar-groove, just above the claw of the petal, is not 

 covered in by a similar formation, but has its edges in 

 close contact with the ovary. Insects therefore that 

 would reach the nectar must first separate the petals 

 from the ovary by force. 



The closure of the cavity in which the nectar is 

 secreted or stored is sometimes effected by the stigma, 

 but not nearly so freq^uently as by the ovary. In such 

 cases the stigma is of considerable size, and is flattened 

 out so as to cover in the tube of the perianth like a lid. 

 The finest examples of this kind of stigma are fur- 

 nished by the grou^ of gentians known as Cyclostigma, 

 a group represented in our Alps by Q. pv/mila, im- 

 hricata, verna, cestiva, havarica, utriculosa, and nivalis. 

 (Plate I. fig. 37 shows a flower of G. havarica L. seen 

 from above ; and fig. 36 gives a lateral view of its 



