THE PRIMITIVE STREAK OP THE RABBIT. 201 



rabbit would not, in all probability, have been upon the lines 

 on which it is — provided a rabbit could ever have been evolved 

 under other circumstances. What I mean is : a bird or a reptile 

 has a large egg because the embryo obtains its nourishment 

 from yolk, the presence of which causes the egg to be so large. 



A mammal does not require the yolk as a nourishment, and 

 therefore the egg is small. A mammal is enabled to develop 

 within the body of its mother because, we presume, originally 

 the embryo was protected within the membranes of a large egg 

 within the oviduct of the mother, and it was only by the 

 substitution, no doubt very gradual, of (i) placental nourish- 

 ment instead of the yolk, and (ii) fluids exerting considerable 

 pressure on the uterine walls instead of yolk, albumen, and 

 shell, that it was possible tor a mammal to dispense with 

 a large-yolked egg and to be developed under the conditions 

 in which we now find it. 



Thus it is to the phylogeny that the conditions under 

 which development takes place to-day are due. Now these 

 conditions were imposed upon the development at a compara- 

 tively very recent period in the evolution of the rabbit. 



So when we find that certain forms assumed by certain 

 centres of activity, which centres of activity themselves 

 undoubtedly date back to infinitely earlier epochs than the date 

 of these superimposed conditions, are due entirely to the con- 

 ditions of this day, we must be very careful in drawing any 

 morphological conclusions from those forms assumed. If, for 

 instance, I can show that the changes in form of the primitive 

 streak from a nearly circular spot to a linear expression, and 

 its groove which appears during part of its existence, are due to 

 these present conditions alone, then to draw any conclusions 

 such as frequently are drawn — as, for example, the theory that 

 the line-like primitive streak and its groove represent the lips 

 and aperture of an elongated ancestral gastrular mouth, is 

 impossible. 



If the blastodermic vesicle, at the time that the secondary 

 area of activity is established, became an object freely suspended 

 in some fluid, and was under no influence of internal hydro- 



