CROSS - AND SELF-FERTILISATION 57 
again crossed, and so on for a number of years. The results 
were surprising. Even in one generation the offspring of 
cross- were superior in height, weight and fertility to the off- 
spring of self-fertilisation ; only a few exceptions occurred. 
In the succeeding generations the result became more and 
more marked. We may thus conclude, not, as has been so 
often done, that self-fertilisation is necessarily or in itself 
harmful, but that in general the offspring of cross-fertilisation 
will vanquish that of self-fertilisation in the struggle for 
existence, if the two be brought into direct competition, 
other things being equal, and that in general cross-fertilisa- 
tion is much to be preferred, though self-fertilisation is 
better than none. 
The advantage due to crossing shown in Darwin’s results 
is compounded of several factors, fertility of the parent 
plant, strength of constitution of the offspring, and fertility 
of the offspring. Besides the advantages thus measured, it 
is probable that an important gain is made in the increased 
variability of the offspring resulting from cross-fertilisation. 
On the other hand as compared with self-fertilisation 
crossing has certain drawbacks. It is (1) much less easy 
to ensure fertilisation when the pollen has to come from a 
distinct plant, and (2) fertilisation tends to be delayed (a 
point of importance in the short summer of alpine and 
arctic regions, where seed must be ripened before winter). 
The advantages of cross-fertilisation are often great, and 
frequently enormous, and as at the first glance they appear 
to be obtained at little or no cost, we are inclined to expect 
this method of propagation to prove almost universal, and 
in fact it is extremely common. Self-fertilisation, however, 
is certainly the rule in many plants, although as the flowers 
are open there is a remote chance of a cross ; in a few cases, 
so far as we know, the flowers never open, and crossing is 
absolutely prevented. 
An important conclusion easily drawn from the facts is 
that cross-fertilisation costs the plant a certain price. If 
self-fertilisation be the rule, there is no necessity for any 
further complexity of flower beyond the organs bearing 
pollen grains and ovules (i.e. stamens and carpels), except 
something to ensure that the pollen shall reach the stigma, 
and be protected from injury by rain, &c. No more pollen 
