425 µm or smaller. Grab samplers are lowered to within 3 m of the streambed, stopped, and then allowed to drop to the streambed. This process helps minimize disturbance of the substrate by the descending sampler and achieves a more uniform depth of substrate penetration. Recovered grab samplers are carefully checked to make sure that sample material was not lost because of rocks, sticks, or other debr is catching in the jaws of the sampler. Individual grab samples are composited in a suitable sample container, such as a 19-L (5-gal) plastic bucket or a large dishpan. Each of the samples to be composited is taken within the same instream habitat type but at sufficient distances apart to avoid interference among samples. Simple sonar units, such as those used by sport fishermen to locate schooling fish, are used to coarsely approximate stream bottom conditions and depth, to help define the limits of the instream habitat type, and to determine the depth at which the sampler is released.
In large, sandy-bottomed rivers, the faunistically richest instream habitat type may be woody snags or macrophyte beds associated with the margins of riffles or runs, and the appropriateness of these habitats should be evaluated prior to committing resources to quantitatively sampling deep, sandy substrates. The selection of the appropriate instream habitat type for RTH and DTH sampling is based on national guidance supplemented with information derived from the retrospective data analysis and reconnaiss ance sampling or obtained by consulting the study-unit liaison committee, the regional biologists, and the North Carolina Ecology Group.
Special attention is given to sampling woody snags and macrophyte beds in large, sandy-bottomed rivers or streams with otherwise unstable substrates. Under these conditions snags and macrophyte beds can support the faunistically richest communities of or ganisms within the sampling reach. Snags are sampled by cutting off sections of tree limbs with a saw or lopping shears, removing the limb from the water, and collecting the attached invertebrates by hand-picking and brushing the limb surface and cavities. Losses of mobile or loosely attached organisms can be minimized by placing a net (for example, a D-frame dip net or the Slack sampler) downstream from the limb to catch dislodged organisms or by placing the limb in a specialized snag sampler (Thorp an d others, 1992). The Thorp and others (1992) snag sampler clamps the limb between an upstream and downstream net that minimizes the loss of mobile organisms while the sample is cut away from the main body of the snag. If the limb is too large to cut, it is sampled in place by brushing its surface with a fingernail brush and catching the dislodged invertebrates in a net placed immediately downstream from the snag. The lengths and diameters of the limbs sampled should be entered on the field data sheet.
Macrophyte beds can be sampled with disturbance-removal samplers (Slack sampler, grab, or stovepipe samplers). Net samplers, such as the Slack sampler (fig. 5), can be used if there is sufficient current to wash the dislodged plant and animal material in to the net. A knife or trowel is used to dislodge the plant material from the substrate. Grabs can be used to sample low-growing submergent macrophytes on sandy or silty substrates in areas with low current velocities. However, stovepipe samplers may prove more effective and should