﻿EXISTING AND FOSSIL CYCADS COMPARED. 



2 33 



bundle, can readily conceive of an aborting of one of the two seeds with the further 

 final reduction of the whole carpophyll to a simple stem bearing a single seed ; 

 that is, to a moiiosporangiate pedicel. 



The fact has already been pointed out in detail that even in the existing cycads 

 there are concrete examples of concentric bundles which may be arranged in a 

 series possessing all the bundle elements and illustrating the theoretical changes 

 involved in the final phases of reduction to the cycadeoidean form of sporophyll. 

 The outlines of this series, being of such fundamental importance to theories of 

 reduction, are here repeated. 



(a) The concentric megaspore bundle trace of Cycas revoluta with the most 

 leaf-like cycad sporophyll known is highly complex and gives rise to frequent 

 branches containing much transfusion tissue. 



(b) The more reduced megasporophyll of Encephalartos has much simpler 

 megaspore traces than Cycas. 



(c) There is present in the stalk of the megasporophyll of Bowenia a reduced 

 concentric bundle strikingly like that of the seed pedicel of Bennettites, with the 



Fig. 137. — Archajopteris. Replacement of leaflets by grouped sporangia in a Devonic fern. 

 (Frond '4 natural size, and sorus enlarged. From Schimper.) 



endodermis and cortex excluded, as they should be in such a comparison. If, how- 

 ever, comparison be made with the iiitersemiual scale bundle the agreement with 

 Bowenia is quite complete. 



(d) A further example, even more strikingly like the cycadeoideau pedicel 

 bundle, is to be seen in the staminate cones of Stangerict. This concentric type 

 of bundle is therefore of frequent occurrence in cycads. It is, too, especially to be 

 noted in connection with the present bundle series that the single supply bundle 

 of the cycadeoidean megaspore spreads out to form the chalazal region by increase 

 of its transfusion elements and assumption of a perfectly concentric form. The 

 retained endodermis is primitive, while the outer cortex of the pedicel may be 

 considered as the last trace of the ground tissues of the former peduncle, and 

 perhaps identical with subepidermal bast cells. In the interseminal scales, however, 

 while an endodermis like that of the pedicels is not present, an epidermal sheath 

 appears outside of the cells, forming a less conspicuous cortical zone than that of 

 the pedicels. All of these reductions are in the strictest agreement with the phys- 

 iologic requirements of arrangement in a compacted ovulate cone so largely made 

 up of much lignified elements as is that of Cycadcoidea. 





