54 ROYAL SOCIETY OF CANADA 
had shown it to be likewise a feature of the genus Dadoxylon, and | 
this led the way for Grand Eury to establish the real connection be- 
tween this genus and Cordaites; and thus we have come to employ this 
last name for those woods with an undoubted Araucarian structure, 
which have been shown to be related to the leaves and fruits for which 
the name was first employed. Ranging from the Devonian through 
the Upper Paleozoic, into the Mesozoic and Tertiary, there are many 
plants known through their wood, which present many points of simi- 
larity to—we may indeed say of affinity with—the Araucarian type, 
but there are many considerations which lead to hesitancy in regarding 
them as generically identical, and among these we may consider the 
distribution in point of geological time. Thus while Cordaiteæ range 
from the Devonian upwards into the Mesozoic, plants with a well 
defined Araucarian habit of growth and fructification, range from the 
Tertiary down as far as the Jurassic, below which they are unknown. 
In the Paleozoic also, there are a number of plants known by their 
wood only, the exact relations of which are matters of doubt—they can- 
not be referred to Cordaites or with certainty to any other known genus. 
Kraus* has endeavoured to solve the difficulty by referring all such to 
the genera Cedroxylon, Cupressoxylon, Pityoxylon, Araucarioxylon, 
Pissadendron and Protopitys, and there are structural grounds for the 
retention of certain elements of this classification, while admitting the 
general desirability of the principle of condensation. On the other 
hand, Felix and others working on the lines of the present acknow- 
ledged position of Cordaites, and having reference to the geological dis- 
tribution of these various plants, have suggested the propriety of con- 
fining the Cordaitez to Paleozoic time, while Araucarioxylon is reserved 
for Mesozoic and Tertiary times. Under this ruling Dadoxylon be- 
comes a provisional name reserved for all those forms which have no 
recognised affinity. This idea has been recognised and utilised by 
Knowlton,’ and although there are obvious objections to it, it may 
nevertheless be regarded as a good provisional basis for further studies. 
It certainly has the merit of giving a definite line of demarcation be- 
tween recent types and those of more ancient date, when mere struc- 
tural resemblances would tend to identify the two. It will also do 
much toward reducing that endless confusion which comes from a 
useless multiplication of names for plants which, if not generically 
identical, are nevertheless closely allied, and as Knowlton very well 
points out, the name of Dadoxylon does not commit one to a final ex- 
1 Schimper’s Pal. Veg., II., 1874, 369. 
? Proc. U. S. Nat. Mus., 1890, XII., 601. 


