408 THE FERTILISATION OF FLOWERS. [parr m1. 
Treviranus supposed that Polemoniacee fertilise themselves 
before the flowers open (742). 
Gilia pulchella, Dougl., and G. (Leptosiphon) micrantha, Steud., 
are believed by Darwin (167) to be heterostyled. 
Professor Asa Gray considers Paloz subulata, L., to be hetero- 
styled, but Darwin (167) shows that the great variability of the 
stigmas and pollen-grains makes this a perplexing case. He says, 
Possibly this species was once heterostyled, but is now becoming 
sub-dicecious, the short-styled plants having been rendered more 
feminine in nature.” (Forms of Flowers, p. 121.) 
Orv. BORAGINE ZL. 
Tribe Cordiece. 
Cordia, L., is heterostyled, but the stamens are of almost the 
same length and the pollen-grains are of the same size in the 
flowers of both forms (Darwin, 167). 
Tribe Boragee. 
300. SYMPHYTUM OFFICINALE, L.—A white, annular ridge at 
the base of the ovary secretes honey, which is lodged in the upper 
part of the inverted corolla. The whole corolla is 14 mm., its 
upper narrower portion 8 mm.,long. The long, trihedral invagina- 
tions of the corolla, which pass from the boundary between the 
wide and narrow parts into the wide part of the bell, closing in the 
spaces between the stamens, were thought by Sprengel to guard 
the nectar from rain; but as in the preceding species, this service” 
is performed by the inverted position of the corolla. Their use is” 
probably to cause an insect to thrust its proboscis between the 
closely approximated anthers, and not into the wide intervals be-— 
tween the stamens. The sharp teeth which these appendages of — 
the corolla bear aid in this object; they certainly do not act as — 
“pathfinders,” as Sprengel thought, for to an insect inserting its — 
head into the flower from below, they appear not bright and 
shining, as Sprengel describes them, but as dark points, guarding ~ 
all but the legitimate access to the honey. This view is confirmed 
by a comparison of the lengths of the proboscides of insects which ~ 
reach the honey of this flower in the normal way, and of those — 
which do so by biting a hole in the narrow part of the corolla. To 
? 
reach the honey by passing between the anthers an insect requires — 
a proboscis at least 11 mm. long; to reach it by passing between 
i 
J 
