GYMNOSPERM^. 63 



winged seeds. The two principle genera are entirely confined to the Southern Hemi- 

 sphere ; the others have each a single species, and are natives the one of Japan and the 

 other of China. 



The earliest traces of distinctly Gymnospermous wood were for years supposed to pre- 

 sent the Araucarian structure. The foliation of the Permian genus Walchia and the 

 Triassic Voltzia resembles that of Araucaria, but the latter possesses an abnormal cone, 

 <;onsisting apparently of altered and expanded leaves, a peculiarity which is even more 

 striking in the aUied Glyptolepis. Another protean genus, Palissya, seems to 

 approach Cunninghamia. These remote genera are probably, however, extinct types, 

 ancestral perhaps to several existing tribes of the Coniferas. The first distinctly Arau- 

 carian genus Pacliypliyllum is met with in the Lias, and believed by Saporta to combine 

 characters of Ayathis, Cunninyhamia, and Araucaria. In the Stonesfield, Somerset- 

 shire, and Yorkshire Oolites, cones with every character of true Araucaria make their 

 appearance,^ and have been described by Carruthers ; and from Solenhofen by Th. Dyer > 

 while forms agreeing closely with these have also been found in the Jurassic rocks of India. 

 It is not yet known from Cretaceous rocks, for the larger fossil cone figured by Heer as 

 Araucarites NordensJcioldi, from the Upper Cretaceous of Spitzbergen, is a very indistinct 

 coaly mass, and, as he suggests, probably Cycadaceous.^ 



Small Araucaria-like foliage, both of the imbricated and needle-leaved types, is very 

 abundant in the Cretaceous series of Aix-la-Chapelle, and is also met with in several 

 localities on the same horizon in France. It appears probable, from the constant associa- 

 tion of the two types of foliage together in so many places, that they may have belonged 

 to the same plant. 



Of the other existing genera, Cunninghamia, a native of China, has been described 

 from German Cretaceous rocks as Cmininyhamites, and a fragment has even been deter- 

 mined from the Miocene. No fossil representative of the Japanese Sciadopitys is 

 known, except from the Amber-beds of Samland, in Prussia. 



Genus — Agatuis. 



The flowers are monoecious or dioecious, cones globose, scales numerous and per- 

 sistent, the seeds unattached, solitary, or rarely in twos, and occasionally winged on one 

 or both margins ; leaves sub-opposite, ovately oblong or rarely lanceolate, leathery, and 

 finely parallel-veined. 



1 Lesquereux believes that cones of true Araucaria occur in America as far back as the Trias, 



2 'Flora foss. Arctica,' vol. iii, pi. xxxvii, p. 12G. Heer acknowledges the figure to be much too 

 distinct, and that the position and arrangement of the scales can only be made out with the greatest 

 trouble. Restored as it is, it possesses no distinctly Araucarian cliaracters, and while no branches of 

 Araucaria are found Cycadaceous foliage abounds in most Cretaceous rocks in high latitudes. 



