R G. Stapledon and T. J. Jenkin 27 



have been those described by one of us in a previous paper i(i6). Com- 

 parisons are made, as far as possible, between figures obtained by the 

 same method; but this is not always practicable when contrasting 

 meadows with pastures. The actual method employed is always stated 

 in the tables ; and elsewhere mentioned in footnotes. 



III. TYPES OF GRASSLAND INVESTIGATED. 



At the outset it is necessary to distinguish between natural types 

 and semi-natural types. 



By natural types are here to be understood those which historical 

 evidence suggests have never been extensively under the plough or 

 manured; and which, if broken or manured at a remote period, have 

 completely reverted to type. By semi-natural types ^ those which 

 have certainly, at one time or another, been under the plough and, 

 at all events, manured during the rotation previous to reverting to 



grass. TT J J 



The semi-natural types may be further classified thus: Untended, 

 those which have been ploughed and probably manured 50 to 100 

 years ago, and then, after yielding crops for some seasons, allowed to 

 revert to grass without any so^Ndng^, and have so remained ever since, 

 in most cases, without any further application of manures. Tended, 

 those which have been down to grass about 20 to 50 years, in most cases, 

 receiving periodic if but slight dressings of manure (at least during the 

 earher years) and probably sown in the first instance ^^th rye grasses 

 and clovers only, or with loft sweepings. 



The following types, details of which are given hereunder, form 

 the basis of the present investigations : 



Natural Types. 



The Board of Agriculture's returns for 1914 gave 21 % of Welsh 

 land as being under mountain and heath vegetation. Much of this 

 is Molinia and Nardus grassland and moorland vegetation which is 



1 "Specific Frequency," the number of times a species (without regard to the number 

 of plants of that species) occurs per 100 readings within a mesh 6" x 6". "Specific 

 Productiveness "-the percentage of edible herbage arrived at by cutting, sortmg, and 

 weighing. "Percentage Frequency," arrived at by the number of plants of each 

 species lev unit of area. Jenkin has, however, used as his unit the number of md.vidual 

 "tillers" instead of individual plants. ^ „ t a ■+\ 



2 These would seem to correspond to the "Artificially Induced Grasslands of Smith 



and Crampton (13). 



3 Excepting possibly some sweepings from the hay lofts. 



