Grass 
These results are distressingly inconsistent. When 
oracles give out uncertain or directly contrary replies, 
can one blame the practical man for paying no attention 
whatever to their commands? Yet it is essential to 
the formation of a good pasture to have the best and 
most nutritive grasses. 
III. Grasses vary very greatly indeed in the vigour 
of their growth and the amount of hay produced. 
Cocksfoot, for instance, has produced 27,905 lbs. per 
acre on rich sandy loam, as well as an aftermath of 
11,910 Ibs. Perennial ryegrass gave 7827 lbs. per acre 
(when dry 3390 lbs.); timothy has given 40,837 lbs. 
(17,356 lbs. when dry); sweet vernal has given 7827 
Ibs. (2104 lbs. when dry); sheep’s fescue has given 
1280 lbs. (on poor soil, Burger). 
Here again there is a great deal of uncertainty, 
for farmers would scarcely accept these conclusions. 
Timothy can hardly produce more than five times as 
much as the universal ryegrass. But there are experi- 
ments (those at Clifton Park), conducted at great expense 
by Mr. Robert Elliot,? which also seem to shatter the 
reputation of ryegrass, 
1V. Method of Growth. —The “tussock” form of growth 
is very common, but to get a really good pasture, it is 
necessary to have a proper admixture of those other 
kinds which throw out runners, These break out of 
the leaf-sheaths and produce, under or over the ground, 
a network of creeping stems from which leafy shoots 
spring up. Meadow-foxtail and smooth-stalked poa are 
good examples of these turf-forming grasses. 
Then again some roots penetrate deep into the soil, 
exhausting the deeper layers, whilst others are surface 
feeders. Rough-stalked poa, ¢.g., will scarcely go deeper 
than 12 inches, whilst smooth-stalked poa will reach 
16 inches, Sheep’s fescue and sweet vernal are surface 
271 
