556 THE FERTILISATION OF FLOWERS. [PART IIT, 
to form a tube, through which the honey has to be sucked. All 
but long-tongued insects are thus excluded (570, 609). 
Lilium croceum, Chaix., is sterile to its own pollen (248). 
Gagea lutea, Schult., and G. arvensis, Schult.—The former is 
proterogynous, but the anthers dehisce soon after the stigma is 
mature. Bees visit both species (590, 1.). 
Gagea Liottardi, Schult.—Anthers and stigma ripen simul- 
taneously, but the golden colour of the flowers and the abundant 
honey attract numerous insect-visitors, chiefly Diptera, so that 
cross-fertilisation must generally occurs (609). 
Fritillaria imperialis, L., is visited by the hive-bee (590, 1). 
395. CoLCHICUM AUTUMNALE, L.—In most cases, the use 
of a long, tubular corolla is to exclude short-lipped insects 
from the honey when they are useless for the work of ferti- 
lisation. But Sprengel has rightly insisted (702, p. 208) that 
in this plant the length of the corolla-tube (which rises directly 
from the bulb) is merely of use in permitting the fruit to lie 
buried in the earth, and indeed hidden within the bulb, and so 
to be sheltered from cold during the winter. The honey does not 
lie in the corolla-tube, but is secreted by the yellow outer surface 
of the lower part of the free portion of the filaments; and it lies 
within grooves of the corolla, covered over by protective hairs. 
The stigmas mature before the anthers (a fact which Sprengel 
overlooked), but remain fresh and capable of fecundation until the 
anthers ripen. If insects’ visits occur in time, cross-fertilisation 
is inevitable, owing to the proterogynous dichogamy of the flower ; 
if they do not occur till late, self-fertilisation also is possible, but 
it is rendered unlikely by the circumstance that the anthers turn 
their pollen-covered surfaces outwards ; if no insects visit the flower 
this movement of the anthers probably prevents spontaneous self- 
fertilisation. 
On the morning of Sept. 19, 1869, at Driburg, I found several 
flowers still closed in their night's sleep, which when opened 
displayed anthers not yet ripe, but stigmas richly coated with 
pollen. Rather later, when the sun came out, I found several 
males of Bombus hortorum busy creeping or flying from flower to 
flower, and sucking honey from the angles between the filaments 
and petals. As they crept into the flowers they touched both 
stigmas and anthers with their forelegs, head, or the anterior part 
of their bodies, dusting these parts plentifully with pollen in flowers 
whose anthers were mature. 
ee 
