388 MR. W. AND MISS A. BATESON ON FLORAL VARIATIONS 
forces which are at work in the production of Variation will 
depend very largely on the precision with which we shall be able 
to answer these questions, and to determine the degree of con- 
tinuity which is present in the process of Evolution. For if, on 
the one hand, the transition from form to form shall be found to 
occur by insensible and minimal changes which are so small that 
no integral change can ever be perceived, we should recognize 
an analogy with the continuous action of mechanical forces; but 
if it should appear that the series is a discontinuous one, and 
that there are in it lacune which are filled by no intermediate 
form, the analogy would rather hold with the phenomena of 
chemical action, which is known to us as a discontinuous process, 
leading to the formation of a discontinuous series of bodies, and 
depending essentially on the discontinuity of the properties of 
the elementary bodies themselves. 
It may be observed at this stage that in proportion as the 
process of Evolution shall be found to be discontinuous the 
necessity for supposing each structure to have been gradually 
modelled under the influence of Natural Selection is lessened, 
and a way is suggested by which it may be found possible to 
escape from one cardinal difficulty in the comprehension of 
Evolution by Natural Selection. 
For there is one obvious consideration which makes it difficult 
to suppose both that the process of Variation has been a con- 
tinuous one, and also that Natural Selection has been the chief 
agent in building up the mechanisms of living things. This 
difficulty, which is well known, may be stated thus. If the pro- 
cess of Variation is supposed to have been continuous, it cannot 
be supposed that the mechanism was at all periods of its evolu- 
tion so beneficial as to be selected. For, from our knowledge 
of Natural History, we are led to think that while certain devices 
and structures may be beneficial to their possessors, yet they are 
so only by reason of the degree of perfection in which they 
exist ; and that if they were materially less perfect, their utility 
would cease. Besides, even if there had been at some phases 10 
their state of imperfection other functions for which they were 
adapted, yet still in any process of continuous evolution there 
must be substantially many transitional forms which are useful 
for no purpose, and therefore cannot be selected: in short, that 
the evolution of a special contrivance for adaptation is not com- 
