70 NATURAL SELECTION FROM 



tion in certain Gymnosperms, and is a constant 

 feature of Angiosperms. 



It may be said that in all these cases we are 

 dealing with structures that have ceased to be 

 useful, and therefore are being gradually elim- 

 inated, No one can say how useful they are, but 

 no one can deny that they are functional. But 

 there is a striking illustration of another sort 

 among Gymnosperms. The suspensor is a con- 

 spicuous organ of the embryo in this group, with 

 a development apparently out of all proportion 

 to its usefulness. In fact, it is a most exagger- 

 ated structure, often becoming closely coiled on 

 account of its extreme length. One would sup- 

 pose that this would be the first structure 

 eliminated, or at least curtailed, if usefulness de- 

 termines suppression. But the suspensor of 

 Gymnosperms shows no symptom of suppression 

 throughout the whole group, and still among the 

 heterosporous Pteridophytes below and the An- 

 giosperms above, where the same conditions pre- 

 vail, it shows no such imusual development. 



Several illustrations could be taken from Gym- 

 nosperms, all of them fundamental in the struc- 

 ture and progress of the group, and none of them 

 j in use by taxonomists. My claim is that it may 

 I be one thing to pass from species to species within 

 the limits of a small natural group; and a very 

 different thing to pass from one great group to 

 another, I do not doubt that the characters of 

 a genus may have been juggled in a variety of 



