THE ANGIOSPERMAE 



1129 



Without discussing this question at the moment, we must understand that 

 the relationships of the parts on so Hmited an axis as that of the flower can 

 never be exactly sim.ilar to the relationships of the parts on a vegetative 

 axis, even when the receptacle is as elongated as in a Magnolia (Fig. T103), 

 and that the crowding together of parts on the receptacle may give rise to 



Fig. 1 103. — Magnolia pterocarpa, showing the remarkably 

 elongate receptacle. {After King.) 



doubts in some particulars as to the exact limits between axis and appendage, 

 especially in compound gynoecia. The entire surface ot the receptacle is 

 mapped out between the various appendage organs, which originate in very 

 close contact, and even the apex may become, at least secondarily, involved 

 in carpel formation. This is a different thing from stating, as some do, that 

 a single carpel may be terminal on the receptacle. We have seen in the 

 preceding section that in many apparently terminal carpels the vascular 

 supply in the carpel base shows that the apex of the receptacle still exists 

 and has at least some traces of its own vasculation, but that it has become 

 embedded in the base of the carpel and lost to sight. The same is true of 

 some coenocarpous ovaries, as we shall see later (see p. 1229). 



The receptacle is sometimes called the torus. This is incorrect, for the 



