256 GINKGO. 



1894. Ginkgo digitata, var. integriuscula, Bartholin, Bot. Tidss. vol. xix.. 

 p. 96, pi. iv. fig. 1. 

 G. SiMoni, ibid. p. 97, pi. iv. figs. 2 and 3. 

 1897. Ginkgo Sutloni, Bartholin, Dan. Geol. Anders, pi. B, fig. 11. 



G. digitata, Nathorst, Flor. Spitzbergen, p. 15. 

 1900. Ginkgo digitata, Seward & Gowan, Annals Bot. vol. xiv. pis. ix. and x.. 

 G. digitata, Seward, Manchester Lit. and Phil. See. vol. xliv. p. 23^ 



pi. ii. fig. 5. 

 Cf. G. polaris, Nathorst, Norwegian Polar Expedition, pt. iii. pi. i.. 

 fig. 8, etc. 



Type-specimens. The type of Brongniart's figures is probabl}^ 

 in the Paris Museum. The type-specimen named by Phillips- 

 SpJienopteris latifolia is in the York Museum. 



Leaves with a long slender petiole, slightly expanded at the base 

 and grooved on the upper surface, attached to either long or short 

 shoots. The lamina may be either entire or repeatedly lobed, 

 varying in shape from a broad fan-shaped form, with a straight 

 base, or with the lower margin of the lamina making an acute angle- 

 with the petiole (e.g. Fig. 1, PI. IX.), to a narrower wedge-shaped 

 form, in which the lower edges of the lamina make an obtuse angle- 

 with the petiole (Fig. 10, PI. IX.). Veins numerous, spreading,, 

 and branched dichotomously. The male flowers are similar in form 

 to those of the recent species, the filaments of the anthers bearing^ 

 two or more terminal elliptical or oval pollen-sacs. The female 

 flowers are imperfectly known, but probably similar in type to those 

 of Ginkgo hiloba. 



In the above diagnosis, which is necessarily incomplete, the 

 flowers are not fully described, because we have no absolute proof 

 of the actual connection of certain species of Ginkgo leaves with 

 associated male flowers and seeds or fragments of female flowers.. 

 There is, however, every reason to believe that the flowers of some- 

 at least of the Mesozoic species of Ginkgo agreed in essentials with 

 those of the recent species. 



It will be seen from the above synonymy that I have included 

 under Ginkgo digitata certain species which have usually been 

 regarded as distinct. The chief reason for this diminution in the 

 number of the specific names is to be found in the marked 

 tendency to variation in leaf-form of the recent species. The- 

 deeply lobed lamina, such as that illustrated in Figs. 2 and 10, 

 PI. IX., has generally been regarded as characterizing a separate 

 species, named by Sternberg Cyolopteris Suttoni, and referred to by 



