JAN. 1907. 
THE ORIGIN OF SPECIES. 
11 
more we rely upon the testimony of casual observers who merely 
report what they stumble upon without knowing the requirements 
of thorough investigation, the more ''discontinuities" and myste- 
rious saltations we encounter. One continuity properly demon- 
strated, outweighs myriads of mythical discontinuities. 
It is well to bear in mind always that there is no evolution of 
organisms apart from the development of individuals. The evolu- 
tion of a species is but the summation of achievements in individ- 
ual development. Consequently, if we would know whether evolu- 
tion is discontinuous or continuous, we must get our direct evi- 
dence from development as we now find it going on. If in any 
series of developmental stages, we find a single stage that is not 
causally connected with preceding stages, we may pause and ask, 
what is the use of investigating amid miracles of discontinuity? 
In embryological development we find continuity and order of 
sequence regnant throughout. There are no kaleidoscopic leaps 
to interrupt the sequence. The course followed is, in the main, 
the one anticipated in the hereditary foundations of the germ. The 
flow of events may swerve this way or that, but nowhere is there 
a gulf breaking the connection between antecedents and conse- 
quents. 
Of course species are sooner or later separated by gaps. But 
are these gaps present ab initio? The claim of initial discontin- 
uity is usually founded on so-called "sports," the preparatory 
stages of which were overlooked, or very imperfectly investigated. 
This is not the kind of evidence on which to establish a general 
theory. If discontinuity or mutation were the law in the origin 
of species, continuity should not be expected in a single case, as 
De \'ries has clearly perceived. Hence he has taken care to deny 
the possibility of getting new species through ordinary variation. 
In such variation continuity is undeniable. Hence the whole reli- 
ance is placed on ''Sports," and "improved races," however true 
they breed, are not admitted to the ranlc of species. 
We see then how very important it is to make our investiga- 
tion of the origin of species exhaustive. We have to select well 
our material for study, and then attend closely to some one char- 
acter at a time. This means that we must work with living forms, 
which we can keep under unbroken control from generation to 
generation. 
Then we must not stop here, but extend our observation over 
