IMPACTION ON CYLINDERS The efficiency of impaction on vertical sticky cylinders is increased by: (I) increasing wind-speed; (2) increasing the mass of the particle (by increase in size or density); and (3) decreasing the diameter of the cylinder (except that large particles, such as Lycopodium spores, tend to blow off narrow cylinders at high wind-speeds, whereas small spores, such as those of Ustilago perennans, do not). As will be shown later, similar relations hold for impaction on surfaces of other shapes, such as spheres, disks, strips (ribbon) and—in the field—leaves, stems, and stigmas. The value for E depends also on the shape. The order of decreasing efficiency is probably: spheres, disks, cylinders, and strips (other conditions being equal, including radius or half-width).
Wind-tunnel results confirm the conclusion of C. N. Davies & Peetz (1956), that under extreme conditions efficiency of impaction can be zero; for example the very small spores of the puffball, Lycoperdon (Calvatia) giganteum, were not impacted under any of the conditions tested. The slightly larger spores of the smut fungus, Ustilago perennans, however, showed appreciable efficiencies—but probably because they often travel through the air in clumps which effectively behave as heavier particles than their component single spores.
The deposit is not uniform over the cylinder, but is densest on the part of the circumference projecting farthest forward into the wind; this is the so-called `stagnation line'. The density of the deposit falls off towards the sides of the cylinder, and there is a spore-free zone on the shoulders where the air-flow is more or less tangential to the surface. Usually there is no deposit at the back or downwind side of the cylinder, but with Lycopodium spores (32 it diameter) a narrow line of deposit has been observed down the back of cylinders of less than o•1 cm. diameter at wind-speeds of i metre per sec. or less. Deposit on the side opposite to the oncoming wind is negligible or zero under most conditions tested. The angle subtended by the deposit on the upwind side was less than in both turbulent and streamline air. Other things being equal, the angle subtended at the centre of the cylinder by the deposit was increased by increasing the wind-speed and by decreasing the cylinder diameter— increasing efficiency of the whole cylinder evidently runs parallel with increase in the angle subtended by the deposit. The low efficiency of wide cylinders and slow winds shows both as a narrower trace, and thinner deposit per unit area.
The decrease of efficiency with increasing cylinder size was first noticed in field tests. Per unit length, a large cylinder 12 cm. in diameter may collect no more pollen grains or spores than a cylinder i cm. in diameter, and per unit area of surface it may collect many fewer (Gregory, 1951).