18998) ANNULATE ANCESTRY OF THE VERTEBRATA 19 
of any group of animals, we must, if possible, discover the method 
of feeding which caused the ancestral form to depart from its 
congeners. Strong in this conviction, therefore, I made a special 
comparative study of the Arachnida, extending over four years, and 
ultimately endeavoured to show! that their peculiar morphology could 
be explained in detail as an adaptation to their method of feeding. 
It was therefore only natural that any suggestion, even the 
faintest, as to what might have been the primitive method of feeding 
of the ancestors of the vertebrates should lead me at once to see 
whether the same principle could not be made to apply again. 
Was it not possible to deduce the typical low vertebrate from a 
hirudinean by a series of structural modifications resulting from a 
further development of the leech-method of feeding? No one will 
deny that such a possible solution was worthy of investigation, 
even though he may not be so convinced as I am that morphology 
is an aimless pursuit unless it go hand in hand with physiology. 
Although most zoologists admit this latter as a pious opinion, in 
practice it is too often ignored, as may be gathered from the fact 
that every attempt to discover the ancestry of the Vertebrata, 
with which I am acquainted, has been based solely on structural 
similarities, while the functions of the structures themselves have 
been treated in a most arbitrary fashion. For example: the 
central nervous system of the Arachnida is said to have be- 
come the central nervous system of the Vertebrata, with an 
entirely different organism to be innervated. The intestine of the 
king-crab is said to have been lost in the spinal cord of the 
Vertebrata, and a new one has been provided; the sheath of a 
protrusible proboscis is turned into the vertebrate notochord ; old 
mouths may close and new ones open, and so on, Continuity of 
function is apparently of very secondary importance, while similarity 
of structure or of mere position relative to other organs is of prime 
importance. I do not call this physiology and morphology going 
hand in hand. It seems to me more like physiology being dragged 
by the neck, while the morphologist demonstrates the perfection of 
his structural resemblances. Hence, all the arguments to which we 
have hitherto been accustomed have seemed to me from the first to 
be hopeless. It is not alone to the discovery of similarities of 
structure that we must look for clues to evolutionary progress, but 
rather to the development of new functions in response to some 
probably gradual change in the environment, these new functions 
leading, whether by selection or inheritance, or both together, to 
modifications of structure, subject always to the physical laws of 
the environment. 
1 “Comparative Morph. of the Galeodidae.”—Trans. Linnean Soc., vi., pp. 305-417, 
189 
lor) 
