BOTANICAL EVOLUTION, 467 
flower. The moment that this antenna is touched, the pollen which 
is all joined together into one pear-shaped mass and furnished with 
an adhesive disc, is thrown forward with considerable force and 
adheres to the insect, which, alarmed at its rough reception, flies off to 
auotuer flower. The spring is so great that the pollen-masses have. 
been thrown to the length of three feet and more. 
in all these remarks) which I have made on the modes of pre- 
venting cross-fertilization I have been referring to flowers which in 
structure are hermaphrodite, though they are not so in function, 
But in many plants the process has gone a stage further, and by the 
abortion or total suppression of one of the sets of essential organs 
a complete separation has been effected, so that only cross-fertilization 
can take place. In some cases both kinds of flowers are formed on 
the same plant, one kind producing stamens only and the other 
pistils only ; such plants are termed monecious ; a familiar example 
is the Vegetable Marrow. But the furthest degree to which this can 
go, is when these different kinds of flowers are produced on different 
plants. In such cases of course only those plants which produce 
pistillate flowers can have seed, and supposing that they represent 
fifty per cent. of the whole, then the other fifty per cent. are sterile 
and produce no seed. I have already stated that the supreme object 
of the flower is the production of seed, therefore in these diacious 
plants we may be sure that the giving up of their seed-producing 
power by one-half of all the plants of any species, simply means that 
the cross-fertilization thus secured to the other half more than 
counterbalances the apparent first loss. For we cannot conceive of 
any such variation arising and being intensified in so many cases as it 
is, were it not advantageous to the race. In our New Zealand flora, 
we have singular facilities for studying this question of the separation 
of the essential organs in flowers, for in proportion to the total 
number of our species we have a larger number of dicecious and uni- 
sexual forms than probably any other known region. The causes of 
this I am not very clear about, nor do I think we have sufficient data 
to decide on the question until more is known about our insects. 
Three of our commonest flowers show the steps of this process of 
separation very well. Thus our common fuchsia (/. excorticata) has 
two kinds of flowers. The first and most abundant form is usually an 
inch or more in length, is of a green colour more or less tinged with 
purple, has eight stamens whose anthers are full of bright blue 
pollen, and a complete pistil formed of four united carpels. Itisa 
hermaphrodite, and though probably not self-fertilized, yet every 
flower is capable of producing a fruit. In the second form, the 
flowers are individually smaller, and are of a paler and more pink 
colour. They are also hermaphrodite in structure, but a closer 
examination shows that the anthers have no pollen, so that unless 
they are fertilized by pollen from the larger form, they cannot 
produce fruit, so that this second form while retaining the herma- 
phrodite structure has become unisexual in function. I may just 
remark here that the fuchsia and a few other long-tubed flowers have 
become specially adapted for fertilization by tuis, and other honey- 
birds. : 
The next stage in advance is seen in our common large Clematis 
(C. indivisa), in which, also, two different kinds of flowers are 
