INTKODUCTION. XXXI 



and which, moreover, must unavoidably include a full share of 

 the errors and oversights inseparable from a first attempt of its 

 kind, and one brought out under very unfavourable conditions 

 of age and leisure. Doubtless, all botanists of liberal thought 

 will look chiefly to what a first such compilation may accomplish 

 under difiicult circumstances, not to what it leaves non-accom- 

 plished ; but liberality of thought is far from universal. 



In the ' Compendium of Cybele Britannica ' the Index of 

 names, specific and varietal, counts up to 2600 ; of which (say) 

 100 may be only second names for the same identities. But 

 numerous varietal names of ferns and grasses, of brambles and 

 briers, of mints and hawkweeds, and of some smaller genera, 

 do not appear in that Index ; so that we should be counting 

 considerably under the full figure in taking only 2500 named 

 species and varieties for the alleged fiora of Britain. 



Say, that we should not much exceed the mark by taking it 

 at 3000, including all the named varieties, all the yearly 

 increasing casual plants, and all others improperly reported as 

 quasi-wild in Britain. The number of counties, along with the 

 subdivision of the larger counties into vice-counties, amounts to 

 112 ; many of the commoner plants occurring in aU or nearly 

 all of these counties. Perhaps we may fairly take 50 counties 

 and vice-counties, something under the half, as an average 

 number for each plant. Multiply the one figure by the other, 

 3000 by 50, and the result would appear in the fearful number 

 of 150,000 special items or facts, real or imaginary; each of 

 these items consisting of the name of a county or vice-county, 

 along with the name pf at least one botanist selected as voucher 

 or testifier to the fact. 



It is difficult to conceive any method by which these double 

 names could be clearly quoted, without allowing on the average 

 a half line of text to each pair of them. Now, 75,000 lines 

 printed at the close rate of' 40 lines to a page, over and above 

 the space needed for plant-names, would require 1875 pages of 

 text, without reckoning in the extra pages further required for 



