253 



conoluding that ancestral forms of tlie recent plant existed as 

 far back as the Upper Carboniferous period. Since this paper wa* 

 written many new facts have come to light, which enable ^is to 

 extend the geographical distribution of the family and strengthen 

 Heer's conclusion as to the importance of these Gymnospermous. 

 plants in the Mesozoio or even in the Palaeozoic epoch. 



Such genera as Qinhgophyllum, Saportma, Tricliopitys, Dicrano- 

 phyllum, Rhipidopsis, Whittleseya, and even Ginkgo itself, hav& 

 been described from Carboniferous and Permian rocks as probable 

 members of the family to which Ginlcgo biloha belongs. As 

 regards some of these genera, there is hardly sufficient evidence 

 in favour of their iaclusion ia the Ginkgoales ; on the other 

 hand, the close resemblance of the Permian leaves referred to. 

 Ginlcgophyllum, Ginlcgo, Saportcea, and Baiera, to the recent plant, 

 render it probable that closely allied species existed in the 

 Palseozoic era. Certain fossil seeds from St. Etienne, of Permian 

 age, described by Brongniart ' as species of Cardiocarpus, are 

 almost identical in structure with Ginkgo seeds. It is, however, 

 from Mesozoic strata that we obtain the most striking proof of 

 the abundance of Ginkgo-Vike trees in the vegetation of the past. 



Although as a rule it is, for various reasons, preferable to 

 avoid the application of the name of a recent genus to fossil 

 species, yet the generic names Ginkgo and Salishuria have been 

 so generally used for fossil leaves, and on evidence of a trustworthy 

 nature, that it would be inadvisable to suggest a departure from 

 so well established a custom. The leaves described by Gardner 

 from the Tertiary beds between the basaltic lava-flows of Mull, 

 and referred to Unger's species Ginkgo adiantoides,^ agree so 

 closely with those of the recent species that we may well hesitate 

 to admit even a specific diiference. When we come to examine 

 the Jurassic and Cretaceous species, examples are not lacking 

 which also exhibit the closest agreement with the surviving type. 

 It has been pointed out by more than one author that the number 

 of specific names applied to Jurassic Ginkgo leaves is excessive ; 

 palseobotanical writers have frequently overlooked the wide range 

 of variation exhibited by leaves on the same tree of a living 



' Brongniart (81). 



'' Gardner (83), pi. xxy. 



