WIND-TUNNEL STUDY OF IMPACTION The following account is based on work with a small, low-speed wind tunnel built at Rothamsted Experimental Station in 1949 (Gregory, 1951; Gregory & Stedman, 1953) and includes some hitherto unpublished data.
The wind-tunnel consists of a horizontal square duct (Fig. 11). The two ends of the tunnel project through the end walls of a small building which forms a laboratory traversed by the eight-feet-long working section of the tunnel. The tunnel uses outdoor air which is passed through once only and not re-circulated. A four-bladed wooden propeller absorbing o•56 horsepower at 2,85o r.p.m. in the exit draws air down the tunnel. An expansion section converts the 29 cm.-square working section to the 46 cm.-circular diameter at the fan. The flared intake-end is of 51 cm. square cross-section, contracts to a bell shape, and contains a paper honey comb `straightener' to remove eddies and produce streamlined flow; but if turbulent flow is needed a constriction is inserted in the first part of the working section (indicated by dotted lines at t in Fig. 11).
Spores under test are injected or otherwise liberated, usually on the tunnel axis near the constriction. Spore trapping equipment, plants, etc., can be inserted farther downwind through removable panels in the walls of the working section. Wind-speeds of from 0.5 to nearly io metres per second are obtained by changing pulleys on the belt drive between the constant-speed electric motor and the fan, and by inserting screens across the tunnel to increase its resistance. The lower wind-speeds are best obtained by increasing the resistance of the tunnel, rather than by slowing the fan, because outdoor wind movement disturbs the flow less when the tunnel resistance is high than when it is low.
Most tests of spore dispersal or deposition in the wind-tunnel involve knowledge of the time-mean spore concentration of the air. The Cascade Impactor (K. R. May, 1945; see also Chapter VIII), operated isokineti cally (i.e. with the orifice facing the wind and with suction adjusted to draw air in through the orifice at the same speed as the wind) is taken as standard to estimate the mean number of particles per cubic metre of air during the period of the experiment. With this information, and knowing the wind-speed, we can calculate the area dose (A.D. = xut). The trap ping efficiency of any surface exposed to the spore-cloud in the wind tunnel can then be determined by estimating the number collected per square centimetre of trap surface, and expressing it as per cent of the area dose.
Liberation of one million Lycopodius* spores at a point on the central axis of the tunnel produced a conical cloud. At a sampling point on the central axis of the tunnel, 1.4 metres downwind, the area dose was about spores per sq. cm. under turbulent condition at wind-speeds of from 5•75 to 9-7 metre per sec. Under streamline conditions the dispersal cone was visibly narrower and the area dose nearly double at these wind speeds. At i i metre per sec., however, streamline conditions gave a much lower area dose because the cloud was displaced downwards under gravity. On the whole, efficiency of impaction was not much affected by whether the flow was turbulent or streamlined.