IN PLANTS HAVING IRREGULAR COROLLAS. 387 
the results that we see have come about by the occurrence of a 
series of Variations, it is of the first importance to know what 
kind of series is constituted by these varying forms, and on the 
answer to this question ail advance in the study of Evolution 
depends, In the absence of some knowledge of the mode in 
which variations occur, it is useless to guess at the relationships 
or past descent of existing forms; while conjecture as to the 
developments which may in the future be possible to these forms 
is still more hopeless. 
The first question, then, to be answered is this: If we had 
before us the whole series of individuals which have occurred in 
the line of descent of one given form, what kind of series would 
these individuals compose? It is too much to expect that we 
should discover what the series has actually been, for the evi- 
dence is gone; but we may reasonably hope to find out what are 
the general characteristics of such a series, for the series itself 
is still in progress. It is nevertheless a remarkable circumstance 
that a great deal of enterprise and research has of late been 
given up to the object of ascertaining what the actual pedigree 
of divers forms has been, while no one has yet succeeded in the 
preliminary task of determining any of the general characters 
of such a series. Yet if the modern conception of the manner 
of origin of new forms is a right one, it is a process now at this 
time occurring, which by common observations we may now see. 
Surely such observations may become the foundation of a solid 
and reasonable knowledge of the attributes of the method 
of Evolution; and when such a base shall have been estab- 
lished, it may perhaps be safe to attack the special problems of 
descent. 
Supposing, then, that such a series of ancestors were before 
us, the matter to be determined would be the degree to which 
the series is continuous or discontinuous: that is to say, whether 
the differences between any one member and its immediate suc- 
cessor are so small as to be imperceptible, or whether there are 
distinet and palpable differences between them; or whether they 
are sometimes small, and sometimes so great as to cause inter- 
ruptions in the series and to divide it into groups of which the 
composing members are similar, though the successive groups 
are unlike each other. 
The success of any attempt to comprehend the nature of the 
