290 
Canadian Agriculture. 
the day's work fairly started, took the locomotive and his car, 
and went on in advance until he found a suitable place for the 
site of the next farm, when he would return and get the men, 
teams, and outfit together, and transport them during the night 
to the next field of operations. In all, ten experimental stations 
were established : the breaking throughout was found to be 
easy, and the soil in every case good. It was proposed to 
cultivate the ground thoroughly, in accordance with the most 
approved methods of breaking and backsetting, so as to have 
a seed-bed read}' by the spring of 1885 ; those in charge of the 
work were, however, led to believe that a good crop might be 
obtained, even off the sod, and it was resolved to make the 
attempt. Accordingly, on March 27th, 1884, another novel 
train left Winnipeg, taking boarding cars, men, teams, imple- 
ments, and seed-grain to commence the spring sowings. At 
Winnipeg there was still much snow on the ground ; but on 
ffoinff westward the weather became milder and the snow dis- 
appeared, so that when Dunmore was reached, on the 29th, 
the snow was entirely gone, while the ground was dry and 
already thawed to a depth of several inches. The train 
arrived at Gleichen, the most westerly of the farms, on the 
30th, and ploughing and seeding commenced on March 31st. 
Bearing in mind that the sod was not broken till October, and 
that the soil had since then been continuously frozen, it is 
evident that the sod had no opportunity to rot, and the land, 
when spring seeding began, was practically in the same condition 
as when left in the fall, so that a proper seed-bed could not be 
prepared, consequently much of the seed remained on the sur- 
face to be withered by the sun or eaten by birds and gophers. 
To save space, I have condensed the particulars relating to the 
several farms into the Table on the opposite page, the yields 
given in which were ascertained by accurately chaining the 
ground and weighing the produce, this work having been 
entrusted to a qualified Dominion Land Surveyor. 
I have given the yields in the nearest whole number, arid the 
average vield from all the farms was, in bushels per acre : of 
wheat, 21 i : of oats, 44-^ ; of barley, 23 J ; and of peas, 12^. 
The weights per bushel varied at the different farms, in the 
case of wheat from 59 to 63 lbs. ; oats, 36^ to 43^ lbs. ; barley, 
48 to 52 lbs. ; and peas, in the only instance in which they 
were weighed, scaled 64 lbs. 
With two exceptions, the harvest was completed before the 
end of August, while in no case was sowing commenced before 
the end of March. The short time the crops occupied the 
ground is worthy of note, particularly in the case of Dunmore, 
where sowing took place on the 4th and 5th of April, and 
