220 



" desiderata list " was formed for each district. These were arranged 

 in six classes, which were found very natural and convenient in prac- 

 tice ; namely, 1. Arvenses, i. e., corn-field plants; 2. Pratenses, 

 meadow and pasture plants ; 3. Sylvestres, or wood plants ; 4. Sepin- 

 colae, hedge and road-side plants ; 5. Domesticse, plants growing on 

 walls or about houses; and 6. Aquatiles, marsh and water plants. 

 With these lists in hand especial search was now made for the missing- 

 species in their respective habitats, and as they were found, they were 

 struck out. It is obvious that these, escaping as they did the first re- 

 searches, must in general have been of less frequent occurrence than 

 others found at an earlier period ; though some latitude was required 

 for season, or the inconspicuous nature of the plants. 



The data acquired in the course of the investigation of the Flora of 

 the county in question on the above plan, may serve to correct the 

 statement of Mr. Watson (in one of his works, which the writer has 

 not now at hand to quote with precision), to the effect that a single 

 square mile will be found to contain half the species of a county. This, 

 however, will be found to be a considerable overstatement, unless a 

 square mile be selected containing every variety of soil and situation. 

 If it be assumed that a certain number (f) of species are common to 

 every square mile of a county, and that the remainder are uniformly 

 distributed over it, so that every additional square mile should add so 

 many additional species to the Flora of the list ; then if F and F' be 

 respectively the numbers in the Floras of the whole county (contain- 

 ing a square miles), and of any portion of it containing n square miles : 



it may easily be shown that f = — 



a — n 



For the particular county in question this formula gives 502 species 



common to each square mile ; the whole Flora being about 900 : so 



far more than confirming Mr. Watson's estimate. But there are two 



things which entirely vitiate the above calculation. For first, the law 



of distribution assumed makes all but the most common species equally 



rare, which every botanist knows to be contrary to the truth. And 



secondly, the square mile has been assumed to be an average one : 



that is, to contain clay, sand, gravel, limestone, peat, &c, arable, 



pasture, heath, wood, waste, streams, bog, marsh, standing water, &c, 



in similar proportion to the county at large : all of which it is needless 



to say can hardly be found in one and the same square mile. So that 



instead of five-ninths of the Flora being common to every square mile, 



the writer's experience has been that scarcely one-fourth is common 



to every fifty square miles. And his opinion is, that if a square mile 



