THE MAIDENHAIR TREE SEWARD 451 



and not an island as previously supposed. The flora of Grahamland 

 is exceptional among floras of the Jurassic age in the lack of any 

 members of the Ginkgo group. 



Many floras have been described from Greenland, southern Sweden, 

 Germany, Poland, Indo-China, South Africa, and Australia as Rhetic 

 in age: the rocks so called are intermediate in geological position 

 between the Jurassic and the preceding Triassic period; some cor- 

 respond more closely with rocks of the Lias stage at the base of the 

 Jurassic; some are closer in age to the upper members of the Triassic 

 system. One of the oldest leaves, which it is permissible to speak of 

 as a species of Ginkgo, is from Rhetic rocks in southern Sweden: it 

 agrees in the main with the modern leaves both in epidermal structure 

 and in venation. Leaves apparently representing four species have 

 been described from an Upper Triassic flora in South Africa; this is 

 one of many indications of the abundance of trees having foliage 

 constructed on the Ginkgo plan at a stage in the history of the earth 

 when strange reptilian animals were the lords of creation. One of 

 the South African species is almost identical in shape, size, and 

 venation with leaves previously described from rocks of approximately 

 the same age in Virginia. The leaves had a lamina partially divided 

 into linear segments varying in size and reaching nearly 1 foot in 

 breadth. The same species has been found in Upper Triassic rocks 

 of Queensland. The exceptionally rich flora described in a series of 

 remarkable papers by Professor Harris of Reading from material he 

 collected, 1926-27, in the Scoresby Sound district (slightly north of 

 lat. 70° N.) on the east coast of Greenland includes at least six species 

 of Ginkgo. Some of the leaves bear a close resemblance to the 

 maidenhair foliage not only in shape and venation but in the occur- 

 rence of secretory sacs. One of the species is peculiar in having 

 leaves dissected into four lobes with toothed upper margins, a very 

 unusual character. It is certain that trees with various forms of 

 leaf more or less similar in plan to those of the maidenhair tree lived 

 in the latter part of the Triassic period in South Africa, Argentina, 

 Queensland, and the Northern Hemisphere. It is important to 

 remember that the name Ginkgo has been used in the foregoing account 

 for some leaves which should, strictly speaking, be assigned to Gink- 

 goites. This brief and incomplete review of fossil leaves of many 

 ages and from many parts of the world is sufficient to justify the 

 statement that the existing species is the last of a long line of prede- 

 cessors reaching back to the latter part of the Triassic period, a stage 

 in geological history at least 150 million years ago. 



Having followed the story so far it is natural to ask how much 

 farther into the past has the history of Ginkgo been traced? As- 

 suming, as we do, that this generic type did not suddenly appear as 



114728—39 30 



