440 
The Herbage of Old Grass Lands. 
gramineous herbage formed more than 50 per cent, of the total, 
and that in the remaining ten cases the advantage was with the 
miscellaneous herbage. In table (III.) on the same page, showing 
the percentages by weight of species in the gramineous herbage, 
only those species are recorded which were present to the extent 
of at least 5 per cent, of the gramineous herbage. It may be 
seen that Lolium perenne was first in twenty-one cases, Dactyiis 
glomerata in two cases (No. 6, Staffordshire, and No. 9, Derby- 
shire), Agrostis sp. once (No. 16, Wexford), and Festuca ovina 
et var. once (No. 22, Wiltshire). 
The remarkable proportion of miscellaneous herbage in some 
cases, amounting indeed to as much as 89 per cent, on the 
Somerset turf (No. 7), is probably to a great extent attributable 
to the circumstances under which the turfs were allowed to 
grow. Certainly, under the normal treading and grazing of 
stock, such species as Achillea Millefolium, Leontodon sp., and 
Kumex Acetosa, would never make the display which charac- 
terised several of the turfs. An interesting case in point is 
afforded by the Derbyshire turf (No. 9). This yielded as much 
as 70 per cent, of Rumex Acetosa, and, without mentioning 
this circumstance, I wrote to Mr. Gilbert Murray to ask whether 
or not this plant occurs in any quantity on the pasture. His 
reply, dated July 25, is — " There is very little sorrel or sour 
dock noticeable, and as the land is kejDt closely grazed I have 
little opportunity of ascertaining the names of the grasses." 
This quite coincides with the opinion I had formed that in well- 
grazed pastures the bulky miscellaneous species are kept down, 
whilst the freely tillering grasses supply a large ratio of the 
effective herbage. 
The principle of expressing the grasses in percentages of the 
gramineous herbage only has been adopted because it serves 
to bring out in a more significant manner the relative degree 
in which the various gramineous species are developed. The 
remarkable abundance which the figures assign to Lolium 
perenne merely corroborates what had been anticipated from 
constant observations of the turfs during growth. Dactyiis 
glomerata, though a bulky grass and weighing heavily, was 
present on only half the turfs ; and even in its greatest abun- 
dance (the Derbyshire turf. No. 9) it made up only one-third 
of the gramineous herbage. It is significant that not a single 
specimen of Avena elatior was detected in the whole series of 
turfs ; whilst the broad-leaved fescues, of which Festuca pra- 
tensis may be taken as the type, made but (i poor show. As 
regardg the leguminous herbage the preinier positiou is takeu 
