Some Palo:ol)otanical Aspects. 273 



Scotia Carboniferous may include more than one flora, 

 and that when such cases occur, both floras may not be 

 referable to the same stage. 



The sequence and correlations of the rocks in question 

 as worked out by Dr. Ells and Mr. Fletcher on the strati- 

 graphic basis is shown in the following table extracted 

 from the correlation chart in Mr. Fletcher's paper. The 

 correlations by Mr. Kidston and myself are also quoted 

 from the same chart. It will be understood that in adjust- 

 ing palaeontological correlations to the diagrammatic 

 classification of such a chart it is sometimes necessary to 

 restrict them to narrower limits than the author originally 

 intended, while at the same time the conception both of 

 proportion and of emphasis is lost. Thus a tentative or 

 suggested correlation that is limited in its geological 

 range is often quoted instead of the correlation that is 

 positive but of greater latitude. Usually no distinction 

 is made between definite correlations, opinions as to prob- 

 able age, and mere suggestions. Xeither does such a 

 chart indicate the sufficiency or meagreness of the mate- 

 rial on which the correlation is based. 



Canadian Geolo- Ells & Fletcher 



gioal Survey. in Nova Scotia. 



David White. 



Permian or 

 Upper 

 [bouifer 



Permian. ' Upper Car- Union ? 



irons. 



'Riversdale, Har- 

 rington River 

 Coal Measures. Coal Measures.-' and Cordaite Union? 



Shales (St. John 

 \^Devonian). 



Millstone Grit. Millstone Grit. 



Carboniferous /Carboniferous 

 Limestone. \Limestone. 



rRiversdale and 

 I Coixlaite Shales 

 j (Devonian of 

 1st. John, N.B.) 



,^^ , n THortou (Pocono 



r^ ^ -c ,-/- 1 T rHorton (Lower r t~, , 



Carboniierous /Carboniferous I p .1 -c , | or Pennsylvania 



Conglomerate. \ Conglomerate. I c tt. i j, I Waverly. Newei 



lof Lngland . ,, -r-.-i, > , 



° ' vthan Kiltorcan ./ 



Cl 



