REPORT OF TEE CHIEF ASTRONOMER 835 



SESSIONAL PAPER No. 25a 



With respect to the Quilchena flora, there are six species in the Similkameen, 

 one in the Green River group and one in the Kettle river, and if we accept the 

 Horse- Fly River and Coal Gully beds as Oligocene, then five more species must 

 be added, thus making a representation of thirteen species in the Upper Eocene. 

 Against this we have three species in the Fort Union, one in the Porcupine 

 creek and six in the Red Deer river, making ten species of Lower Eocene type, 

 while there is a very strong Miocene contact through Ulmus and Planera 

 oblongifolia. From these facts the argument would seem to be that the facies 

 are decidedly Oligocene rather than Middle or Lower Eocene. 



The second group of localities embraces the numbers 1428, 1430, 1433 and 

 1436 of the 1905 collections, and 471 of the 1903 collections. The plants found 

 to be represented are as follows : — 



Pinus sp. 



Gleichenia gilbert-thompsoni. 



Gleichenia sp. 



Cladophlebis skagitensis. 



Aspidium fredericksburgense. 



Nilsonia brevipiniia. 



Cycadites unjiga. 



Glypcos'trobus europseus. 



Salix perplexa? 



Populus cyclophylla. 



Myrica serrata. 



Quercus flexuosa. 



Quercus coriacea. 



Sassafras cretaceum. 



Leaves of exogens. 



Leaves of endogens. 



Fruit of Exogen (Dorstenia?). 



Undeterminable. 



Of this list, if we eliminate the doubtful reference to Salix perplexa, we 

 find only thirteen species which may be depended upon, but among these are 

 some which afford a very definite indication of age. Inasmuch, however, as 

 locality 471 is somewhat widely separated from the others, and as a special 

 question arises in connection with 1428, it will be necessary to deal with three 

 sub-groups, i.e., 471, 1428 and localities 1430, 1433, 1436. A consideration of 

 previously described floras which may bear some relation to the present, is also 

 essential. They are represented by the Crowsnest Coal basin at Michel station, 

 B.C., the Nordenskiold river in the Yukon territory; the Vancouver and Queen 

 Charlotte islands. Reducing the various floras which may be so compared, to 

 a tabular form, it will be found that the specimens with which we are at present 

 most directly concerned, establish contact with other floras at only nine points, 

 and with respect to only six special groups. None of them can be directly 

 correlated with the Cretaceous at Michel station, the Nordenskiold river, Van- 

 couver or the Queen Charlotte islands. This arises from the fact that in all 

 of these floras the species presented are to a very large extent new, so that there 

 is no overlapping, and they are in the majority of cases extensions of the pre- 

 viously known floras. This is pre-eminently true of Vancouver island. 



25 a — vol. iii — 54 



