244 '^^^ AMERICAN NATURALIST. [Vol. XXXVIII. 



sive revisions. It was furthermore found that, in order to reach 

 correct conclusions in the case of stems which must often pre- 

 sent marked structural alterations arising through the influence 

 of decay and other conditions attending fossilization in its various 

 forms, it was indispensable that there should be a trustworthy 

 means of comparison with existing types, whereby sources of 

 error arising from eliminated structures might be definitely 

 excluded, and the fossil referred with certainty to its nearest 

 relative. The original intention was therefore modified with a 

 view to meeting the requirements of palaeobotanical research. 

 During the twenty odd years these investigations have been in 

 progress, there has been much change in the views held by 

 botanists respecting the significance of anatomical features as 

 affording evidence of descent, and our own studies brought forth 

 facts which gave repeated emphasis of the most positive kind, to 

 the idea that questions of phylogeny cannot be settled either by 

 the morphologist in the narrower sense or the physiologist, 

 either separately or combined, and that a proper historical point 

 of view can be gained only when to such labors we join the data 

 derived from a critical study of the stem structure in all its 

 details. As the greater may always be held to include the less, 

 the present discussion is to be regarded primarily from the bio- 

 logical point of view, and questions of descent will take preced- 

 ence over those of mere taxonomy. 



The original intention was to make a complete study of all the 

 North American w^oods, comprising, as enumerated by Sargent 

 in his report in the Tenth Census of the United States, some four 

 hundred and nineteen species and varieties ; but the great impor- 

 tance of the Coniferas from an economic point of view, their fre- 

 quent representation in the fossil state and their relatively more 

 simple structure, eventually led to their selection as the one 

 group in which initial studies might be prosecuted with the most 

 immediate and profitable results. While the North American 

 species constituted the original basis, various exotic species were 

 added from time to time with the result that our studies as now 

 completed, comprise ninety-two species from North America, 

 twenty-one species from Japan, and four species from Austra- 

 lasia. This extension has proved of great value, not only from a 



