TESTS OF MACHINERY FOR KILLING COTTON WORMS. Aa 
attitude after dragging among or over the plants in turning, and it 
always tends to spring or throw the nozzles back to such positions that 
they deliver a properly-directed spray into the plants. Where torsion 
without bending is desired in these flexile places a rod extending 
through the interior may be employed somewhat as described for cross- 
pipes and nozzle-arms in the special report of the United States Ento- 
mological Commission that has been prepared. The arrangements and 
constructions referred to have been carefully tested this season to cor- 
roborate the results of previous experience. The principle involved 
is simple and practical in its operation, having been tested at Atlanta, 
and again this year at Selma, Ala. 
The flexile nozzle-arms of the Y-shaped trailing forks which were 
originally designed with flexile stems worked satisfactorily thus at- 
tached; but when these fork-arms were tested on a stiffly hung pipe, 
the spring-rod inside soon proved too weak. The strong pendant 
T-torks with curved or sloping side-arms made stiff proximally, and hav- 
ing 3-ply hose for their distal half or two-thirds, stood severe usage by 
all methods, since they were made of stronger tubing and had much 
stouter spring-rods within. The spring-rod in each arm had its distal end 
soldered in ashort piece of tube abutting against the stem of the nozzle. 
Forks of whatever construction will be guided more by the ridges if 
the arms extend in a somewhat upward direction before becoming hori- 
zontal at the ends beneath the plants, as the median part of the fork 
can then sink into the mid-furrow and be guided by its sides. Prob- 
ably nothing better than the pendant Y-forks and T-forks can be de- 
vised for spraying upward through the center of the plant. An addi- 
tional pair of short arms or of nozzles may be used with advantage to 
discharge from near the median line in divergent direction upward 
through the tops of the plants. The simplest plan is to join these or 
the simple eddy chambers directly to the stem-pipe or its extension, low 
down. Such nozzles may be attached side by side, or in what I call a 
tandem gang. This is a series of short tubes coupled end to end, each 
bearing an eddy-chamber discharge. These may be rotated on their 
axes and so are adjustable to different angles. Those who prefer to 
underspray the top of the plant and care less for its base will find the 
tandem arrangement by itself the best device for that purpose for throw- 
ing from the ground, though the forks answer as well when elevated, 
and may also be used beneath the base. 
The eddy-chamber nozzles seem the best sprayers available for ap- 
plying the poison. These nozzles have been tested this season with the 
discharge-hole of various sizes, from one-sixty-fourth to one-eighth of 
an inch diameter. The smaller orifices give the finest sprays conceiva- 
ble. Indeed, with high pressure, the spray vanishes into vapor and 
steam which does not fall, but rises to seek the clouds. From this the 
damp» particles of poison powder must separate and fall. But with or- 
dinary pressure too fine a spray is not attained. 
