336 
and to fatten readily. The same varieties of the cabbage do not 
yield abundant and nutritious foliage and a copious supply of oil 
bearing seeds.”—(Origin of Species, p. ny.) 
On the other hand, when one portion of a limb is 
removed, the remaining portion is strengthened so as to do 
the additional work thrown upon it. 
“According to the interesting observations of M. Sedillot, when a 
portion of one bone of the leg or fore-arm of an animal is removed, 
and is not replaced by growth, the associated bone enlarges till it 
attains a bulk equal to that of the two bones, of which it has to 
perform the functions. This is best exhibited in dogs in which the 
tibia has been removed ; the companion bone, which is naturally 
almost filiform and not one-fifth the size of the other, soon acquires a 
size equal to, or greater than, the tibia.”—(The Variation, vol. ii., 
p. 296.) 
Thus correlated variation is the agent whereby one 
structure of co-ordinated parts is transformed into another 
structure with parts more or less otherwise co-ordinated. 
But there is another cause which probably is quite as 
efficient and far-reaching. I refer to the fact that the 
co-ordinated parts of the structure are the result of a 
process of development ; and in consequence of this fact 
there is a close sympathy between certain associated 
parts, the result being that when one part is modified, 
other parts are unavoidably modified. 
Those parts which are developed from corresponding 
embryonic parts—as the arm, fore-leg, and wing of 
different animqls, or the fore and hind limbs of the 
same animal, parts which are at one period identical in 
their structure, and necessarily exposed to the same 
conditions—tend to vary in the same way. These parts 
are called homologous, and the principle which expresses 
the relation of one to the other is known as homology. 
We have three kinds of homology—serial, represented by 
the centipede ; lateral, represented by the two corres-