GYMNOSPERMiE. 23 



fruits, so far as I can learn,^ the resemblance being so complete in form and size as to render 

 it certain that they are, if not the same, at least closely allied to the existing species. 

 In comparing the characters of the fossil and the recent examples in detail, the con- 

 dition of fossilisation has to be considered. When ligneous fruits occur fossil in the 

 London Clay, the whole of the ligneous tissue is not replaced by pyrites, for an outer 

 and sometimes several inner coats of lignitic matter remain, but the outer skin on drying 

 bursts and rapidly peals off, so that it is rarely preserved in specimens picked from the 

 beach, even when they have not been rolled by the sea. 

 The removal of this outer coat has left these specimens 

 much smoother than the recent Frenela fruit, and in 

 addition they have been water- worn. The outlines of the 

 valves are marked by grooves, due to the greater thickness 



„,,..., ,,.1 . Fig. 6.— Cone of FiG. 1 .—Callitris 



of the ligmtic layer round then- edges, correspondmg with Caiutris End- EndUcheri. 



the thickening in recent Frenelas, and the small scars at ' 



the base appear also to mark the dove-tailing of the valves into the axis, w^hich is partially 

 effected by two twin spur-like projections. The scar near the termination of the valves 

 is similarly due to a small spur of woody tissue. The fruits seem partly hollow, and the 

 seeds, with which the living fruit is filled, cannot be made out ; but the rough division 

 into four cavities, shown by the abrasion of the base in one of the specimens (PI. IX, 

 fig. 2), agrees fairly well with what would appear if a recent cone were infiltrated with 

 pyrites. The removal of parts of the smooth surface from other specimens discloses 

 an inner granular structure, which has probably replaced the seeds. The determination 

 cannot be regarded otherwise than as satisfactory. 



Caiutris Endlicheri is perhaps the most closely related species, and is described as an 

 evergreen bush, with loose subterete branches, crowded with somewhat slender three- 

 edged branchlets, found in the interior of the eastern part of New Holland and about 

 Port Jackson. 



C. Ettingshauseni is rare in the London Clay of Sheppey, about one in 2000 

 specimens, and has not previously been recorded. The cones are closed, as are the 

 greater number of those of the existing species that I have seen preserved in Herbaria. 

 It presents another most interesting example of variation in the number of scales, in a 

 species in which they do not appear to vary at the present day. Pive specimens are 

 figured. 



The North-African type, which has been identified in great abundance in European 

 Eocenes, especially towards the south, is not yet known to occur in England, though 

 the foliage from Bromley, here referred to Libocedrus, bears a great resemblance to some 

 of the foliage from other Eocene localities, referred to Callitris. There is a considerable 



^ The determination was approved by all the authorities at the Kew Herbarium, including Sir J, 

 Hooker and Prof. Olliver, to whom I have submitted it. Other species of Frenela resemble it, but less 

 closely. 



