60 
out; but in the open flower the anthers are tilted backwards 
and the pores then face upwards (Fig. 42 D, F). In the fully 
open flower the pollen tetrads fall in quantities out of the 
anthers without hindrance. 
The style is bent forwards and downwards (Fig. 42 A, B, K). 
The stigma consists of five low protuberances (which appear to 
be somewhat higher than those in the typical form of P. rotundi- 
folia) seated upon 
the apex of the style 
which is flat and 
has a sharp edge 
(Fig. 42 @). 
I have not ob- 
served secretion of 
honey in any flowers 
on the Greenland 
plants; neither has 
SR Exstam in those from 
Nova Zembla nor 
Porrivs in Finland. 
Cross-pollination by 
N insects must be 
of possible occur- 
rence, and indeed 
Fig. 41. Pirola rotundifolia. 
(From Greenland, Tasuisak; Aug. 1896.) 
Shoot with scale-leaves and 3 year’s growths with (visitors in Finland, 
foliage-leaves. (E. W., 1907.) cf. Poprws 1. c.) be- 
appears a necessity 
cause in the fully open flower the pores of the anthers turn 
upwards, away from the stigma, so that the pollen will have 
difficulty in falling down spontaneously upon the latter. The 
Greenland form grandiflora appears, however, to be somewhat 
better adapted for self-pollination than the European Pirola 
rotundifolia, because in the former the distance between the 
pores and the stigma is only from 3 to 5 mm., while it is 6 
