POLLINATION 
61 
other characters which increase the chance of successful 
pollination. There is much risk of damage to pollen from 
rain, for it is necessary to expose it to ensure its being 
carried away by wind. Sporangia as a rule only open in 
dry air, and thus the risk is lessened. The close packing 
of the sporophylls upon the axis also helps to keep water 
from reaching the pollen. 
At a very early period, perhaps, the transition from 
wind-pollination to insect-pollination ( entomophily ) began, 
and from this period the evolution of most flowers went on 
hand in hand with that of insects, and is best studied in 
connection therewith. It is not difficult to imagine how the 
transition may have begun. Pollen is formed in great 
quantities in anemophilous flowers and is a very nutritious 
food. The earliest flying insects would only have very 
short lips, but finding the pollen freely exposed would be 
able to feed on it, and in this way might get into a regular 
habit of flower visiting. Pollen adhering to their bodies 
might thus be carried from flower to flower, and self-pollina- 
tion also might occur. 
Water-pollinated ( hydrophilous ) plants are few, and are 
all probably derived from land-plants (see Chap. III.). 
The Inflorescence. In some of the older types of 
plants with which we have to deal, e.g. in the cones of some 
Lycopodineae, Equisetineae, and Gymnosperms, there is 
simply an axis bearing an indefinite number of sporophylls, 
of one or two kinds. This is a most primitive type of repro- 
ductive shoot, and we can perhaps hardly call it a flower, 
but it is no great stretch of a term to apply to it the name 
inflorescence , as we have already done. In the vast majority 
of cases, however, the reproductive shoot or inflorescence is 
differentiated into, or is composed of, or bears, a number of 
shoots of limited growth (p. 42) termed flowers. It may be 
that from each of the numerous cones of the primitive forms 
one flower was derived, and that the flowers tended to 
become aggregated together, or it may be that from each 
cone a number of flowers were derived (see Coniferae) ; we 
have at present no means of drawing a definite conclusion, 
but it would seem probable that the former view is nearer 
to the truth. 
Very commonly a plant has only one inflorescence, but 
