INTRODUCTION. 
The main object of this publication is to place on record 
certain results obtained by the writer from the study of pasture 
herbage. A few of the present results and others have already 
been embodied in a paper written jointly by Mr. R. G. Stapledon, 
M.A., and the present writer, and published in the Journal of 
Agricultural Science (18), but the greater part of the results are 
now published for the first time. 
METHODS OF EXAMINATION. 
(a) Percentage Frequency.—This method was originally used 
by Armstrong (1), and was adopted by Stapledon (15) in a 
modified form. It has been re-defined as “the number of plants 
of each species per unit area” expressed in percentages of the 
total herbage (18). 
This method has been further modified to some extent by 
the present writer, as follows :— 
(t) The plant unit has been re-defined : 
i. For most plants, the plant unit shall consist of any 
portion of a plant possessing an independent shoot, which 
can be separated from others with at least three rootlets still 
attached to it. Ifa portion possesses six rootlets and it can 
be separated into two in such a way that each portion 
possesses three rootlets, then it counts as two plant units. 
ii. The plant unit of Z+ifolium repens shall be, “ any 
portion of runner up to three inches in length and bearing 
functioning rootlets.” 
iii. The plant unit for Vardus stricta shall be as nearly 
as possible comparable in size to that of Festuca ovina.? 
iv. In a few cases the plant unit used is, “ any individual 
tiller or shoot irrespective of roots.” This definition applies to 
1 Figures in heavy type and in brackets refer to list of references at the end, 
2 This definition was adopted owing to the plant’s peculiar habit. It is 
obviously unsatisfactory, and in future work it is intended to bring this plant, 
in spite of its peculiar habit, under the general definition (i.) 
