92 REPORT OF THE BOTANIST OF THB 
sets of agitators to prevent banking and clogging, and as a fur- 
ther precaution the sides of the boxes were made nearly perpen- 
dicular; but in order to get as much carrying space as possible 
the front ends of the boxes were slanted over the wheels. This 
proved to be a mistake as it favored banking. The front of the 
boxes should be perpendicular. 
The outlet spouts were made uniform in size throughout their 
entire length and stood nearly perpendicular when in use—an- 
other precaution to prevent clogging. To secure uniform distri- 
bution of the mixture in the row each spout had a cone-shaped 
bridge below the outlet. (See Plate VII, Fig. 2.) It is desirable 
to have the mixture scattered along the sides as well as in the 
bottom of the furrow, so that the coverers in passing will draw it 
over the seed. 
We have used this machine for two years with fairly satisfac- 
tory results and certainly with much more uniform results than 
could have been obtained by the hand application of the mixture. 
The important features of uniform and rapid application are fur- 
nished by this machine, but many improvements can be made 
upon it by careful and neat workmanship. The agitators are 
heavy, bungling affairs, with considerable lost motion. The sul- 
phur-lime boxes should be larger and made of light material; and 
the arrangement for opening and closing the slots which regulate 
the quantity of material sown should be made so strong that it 
will not bend when used. It is probable that the manufacturers 
of garden tools could make a machine on the same lines which 
would work to perfection and thus make the sulphur-lime treat- 
ment an inexpensive method of controlling onion smut, 
SOME OTHER EXPERIMENTS. 
EXPERIMENT TO DETERMINE STAGE OF GROWTH AT WHICH INFECTION 
OCCURS. 
Reference has already been made (page 74) to Thaxter’s experi- 
ments, which proved that infection takes. place below ground 
and while the plant is very young;‘also to Sturgis’ experiments 
on transplanting onions, in which it was shown that onion seed- 
