1104 
Journal of Agricultural Research 
Vol. V, No. 23 
to submergence. In this set of experiments the head upstream from the 
weir was made constant, but the conditions downstream were changed 
by stages in the runs from a free fall of 0.5 foot to a submergence of 0.1 
foot. The discharges through this change of conditions remained the 
same within the limit of the experimental error—0.5 per cent. The 
notches were all thin-edged, the cross section of the weir box in every 
case was large enough for full-contraction conditions, and the escape 
channel was wide enough to allow the sheet of water to expand laterally 
after passing through the notch. In none of the tests was the amount of 
submergence small enough to make it possible to determine whether 
the discharge is actually increased with the small amounts of submer- 
