486 



THE AMERICAN NATURALIST 



[Vol. XL 



of LiriocJendron and Sassafras, and the same variability is true of 

 that most interesting rehc of bygone days, the Ginkgo; while 

 numerous instances of similar variations in other genera could 

 be cited. For a number of years 1 have been engaged in collect- 

 ing leaf specimens to illustrate this variability in a number of 

 genera, and I find this task to be a never-ending source of interest 

 in addition to the invalual)le data which it furnishes for the 

 understanding of earlier floras. 



If it be objected that the consideration of geographically widely 

 removed forms as identical leads only to confusion, the answer is 

 — a consideration of the varial)ility in the livino; (\)mj)tonia, tlie 

 unmistakable proof in its preseiu ami past ili>rril>uti()n of its wide 



Brongniart or Heer or T.esf|ucreu\, without a very serious con- 

 sideration of the generic affinities, in fact it can scarcely be said 

 that we have any generic limits in a host of Mesozoic and Xeo- 

 zoic-genera; all these tend to discredit specific distinction basec! 

 on geographical remoteness. 



These considerations lead me to think that the present is not 



litde light cannot be shcil on tli.'ir lii.t(.f\ . AikI ai flu' 



