48 BOTANICAL GAZETTE LJULY 



I ' 



4. There is a close intergradation among the various fossil 



forms. 



5. The more unusual fossil forms are mostly small leaves, and 

 it is among the smaller leaves on shoots and saplings of Lirio- 

 dendron Tidipifera that we find the most diverse shapes, and those 

 forms which most resemble the diverse fossil forms. 



6. The remains of different fossil species would be unlikely 

 to occur associated with each other. 



7. Liriodendron is a monotypic genus at the present time. 



8. The existing tulip tree very probably extends back into 

 the Tertiary period, there being no great climatic change except 

 during the Glacial period, which was survived by all of our exist- 

 ing arboreal vegetation. It is not stretching a point, therefore, 

 to assume that our existing species of Liriodendron might extend 

 still farther back into the Cretaceous ; other plants do (Mag- 

 nolia), and paleontology affords many examples of such per- 

 sistent animal types. 



9. Nearly all of the fossil forms of Liriodendron, if found at 

 the present time, would be unhesitatingly referred to Z. Tulipift^r^ 

 from a consideration of their leaf form alone; and practically all 

 of the abnormal leaf forms of Z. Tulipifera, if found as fossils, 

 would be considered distinct species. 



10. If we assume that in the ancient forms of Lirioilendron 

 the largest and best developed trees were on hillsides, as is the 

 case in the modern tree, then the smaller, more aberrant forms^ 

 which occur on vigorous saplings and shoots, inhabiting the 

 more marshy situations, would be the forms mostly likely to 

 become fossil. 



11. A warm, humid climate during the Cretaceous might 



m 



account for the variety in shape and size of the leaves, just as in 

 Z. Tidipifera the best nourished individuals are the most variable* 



12. The fact that so many of the fossil forms were contem- 

 poraneous militates against considering them as different stages 

 in the development of the genus. We might with equal pro- 

 priety consider the existing varieties as actual stages. 



13. The barren record of the Tertiary period would seem to 



