1 1 8 STUDIES IN FOSSIL BOTANY 



were of great length, extending inwards as far as the 

 axis, parallel to the stalks of the segments, between 

 which they were closely packed (Fig. 49). The sterile 

 segments or bracts were slightly longer than the 

 sporangiophores, so that their laminae overlapped those 

 of the latter. The packing was so close that the fertile 

 laminae were grooved above and below to allow the 

 sterile pedicels to pass through (Fig. 49, Sp.a,f). The 

 sterile lamina was itself a complex structure, for it 

 divided into two apical prongs, directed almost vertically 

 upwards, while it was prolonged below into two shorter 

 and stouter outgrowths (Fig. 50, /, and Fig. 49, st). 

 The whole external surface of the cone was thus 

 effectually protected by this complex system of over- 

 lapping laminae. We see then that the complete 

 sporophyll consisted of three lower sterile segments 

 and three sporangiophores above them, each of which 

 bore four sporangia, so that the sporophyll as a whole 

 produced no less than twelve sporangia. As there 

 were eleven or twelve such sporophylls in a verticil, and 

 the verticils were very numerous, the total production of 

 sporangia by a cone of Cheirostrobus must have been 

 on a great scale. In each sporangium a vast number 

 of spores were produced, which were all of one kind, so 

 far as the single known specimen shows. 



It remains for us to describe the anatomical structure 

 of the strobilus. Its axis was traversed by a central 

 cylinder ; the wood, which is alone preserved, had a 

 stellate transverse section, with about twelve prominent 

 angles, corresponding in number and position to the 

 vertical series of sporophylls (Fig. 51, x). At these 

 angles the spiral elements are placed, so the development 



