76 INTRODUCTION. 
by English writers on questions in phyto-geography. On the 
whole, after eliminating from the Greenland list all the migratory 
weeds and other plants likely carried thither by human agency, 
Greenland may be found to possess a slight excess of American 
species and subspecific varieties; though its floral balance is 
nearly even between the two sides of the Atlantic. And the flora 
of arctic and boreal Europe being more exactly known, than is the 
flora of the corresponding latitude or climate in America, addi- 
tional identities are more likely to come from the latter side. Of 
course, so far as British species alone are concerned, Greenland 
might be treated as au outlier of Europe, wanting some of the 
plants which extend westward to Iceland or Faroe, with milder 
climates. The non-European plants found in Greenland and 
America, are non-British also. 
Au attempt was first made to show the distribution of our 
native plants outside Britain, by stating their limits in latitude 
and longitude, expressed in numerals or the degrees of mathe- 
matical geography ; but the result was found unsatisfactory, and 
was abandoned for the simpler and more intelligible plan of 
naming habitats. Still, the habitats themselves have been so 
selected and arranged that they do indirectly express ranges of 
latitude with considerable accuracy. For instance, in the series of 
names ‘ Channel, Netherlands, Denmark, Gothland, Norway, 
Sweden, Lapland, Finmark,” the range or limit of latitude north- 
ward is approximately indicated. At whichever name in the series 
the northern extension is arrested, the latitudinal limit may be 
inferred with a near approach to exactness. 
These three lines will bring together, in the form of habitats or 
names of countries, a considerable amount of information towards 
showing the areas and ranges of British plants beyond the small 
island of Britain itself, although not carried into the intertropical 
and austral lands. Still, the Author requests his Readers clearly 
to understand, that the epitomized sketch of external distribution 
cannot pretend to be complete in itself, or to Le free from errors 
and oversights. It is impossible to sift such indications of widened 
