air sampling sites should be chosen to represent the range of conditions that might occur. Often, practical considerations require that a single site in each community or in each interior be chosen for all air sampling. Outdoor monitoring is usually done at single sites on rooftops to avoid undue contributions of local sources. Indoors, the site at which a group of people spend a majority of their time (e.g., the living room of residences, open office spaces in large buildings) or at which one or more individuals experience symptoms is usually chosen for air sampling. Material is collected from reservoirs from which exposure can be assumed or extrapolated using mathematical models (Swanson et al., 1990).
Allergen levels in both air and reservoirs change over time as well as in space. Ideally, it should be possible to evaluate air samples continuously over time. As a substitute, multiple discrete samples can be taken over relatively short periods of time, or long-term samples can be taken and analyzed as a time-weighted average or in discrete units.
Sampling for allergens always occurs in a complex environment, and consideration must be given to factors that might modify source strengths, emission rates, accuracy and precision of sample collection, or sample analysis methods and health effects (O'Rourke et al., 1990). Controls are necessary for each step of the investigation.
Sample collection methods include observation, bulk or reservoir sampling, and air sampling. Observational sampling can include sensory perceptions of an indicator (e.g., odors or visible fungus growth) and observation of factors known to be related to specific kinds of sources. (For example, mites are likely found in carpeting on grade-level concrete floors.) Observations can be formalized or casual, the type used by most environmental investigators.
Source or reservoir samples have been used as indicators of exposure to most indoor aeroallergens. Dust samples have been collected for analysis of microorganisms (Gravesen, 1978; Saad and el-Gindy, 1990), arthropod allergens (Ishii et al., 1979; Swanson et al., 1989), and mammalian allergens (Ohman and Lorusso, 1987). Dust collection can be standardized by using a dust collector that maintains a constant flow, vacuuming a given surface for a standard length of time or vacuuming carefully measured sections of a surface, or by using a combination of these methods. A number of measured areas of each substrate can be sampled to estimate variability in space. Vacuum-collected dust samples are weighed before or after sieving to remove hair and other large irrelevant fibers and particles. The levels of allergen per gram of dust can then be used to calculate source strengths