30 PASTURE STUDIES: SOME RESULTS. 
at 54 years to 8-7 per cent. at 9$ years, followed by a decrease to 
68 per cent. at 104 years. 
The grass was much more abundant on Area 1 than on 
Area 2 at corresponding periods. On the former area there had 
been a sudden increase from 4-8 per cent. at 44 years to 9-3 per 
cent. at 54 years. This large and sudden increase may be 
temporary only, but such a change was not met with on Area 2. 
The grass seemed to behave differently also on Areas 2 and 3, as 
on the latter area there was a marked decrease from 84 to 94 
years. On Area 4, however, its P.c. fr. at 134 years was very near 
that for Area 2 at g} years. 
On the whole, therefore, this grass seems to increase fairly 
gradually up to about ro years on this soil, but then acquiresa 
degree of stability. On soil-type i. (Areas 6-8) its developmemt 
seems to depend very largely on the species it has to compete 
with. On Area 7, where its chief competitor was Agrostis tenuis, 
it made very little progress up to 63 years, but on Area 6, where 
A. tenuis was apparently kept well in ckeck by other species, it 
was more successful. 
Agrostis spp. 
Only traces of these grasses were found on an extreme type 
of poor pasture investigated by Armstrong, but on first-rate old 
pastures he found them in quantities up to ro per cent. of the 
total herbage (1). On the best pasture investigated by the 
present writer they had a percentage frequency of 27:1. The 
highest P.c. fr. found was on a neglected 12 year old pasture, 
where it reached g2-9 of the total herbage.? 
Both A. alba and A. tenuis occur on Areas 1-4, but they 
were not separated in the examinations. On these areas, they are 
probably the most troublesome of all arable land weeds, and they 
are very difficult to eradicate even with the most thorough, clean- 
ing. Brenchley (3) obtained 43 “seedlings” from the soil from 
two holes Gin. x 6in. x gin. in arable land “under ordinary 
farm management,” and they were recorded by Stapledon (£7) ia 
2 Great Orme, Llandudno, 500 feet, 
