=132. Choice of stations.= This method is based upon simultaneous readings by means of simple instruments in a series of habitats, or of stations in a single habitat. Such readings are necessary for the variable atmospheric factors, humidity, light, temperature, and wind. Frequent readings suffice for water-content and precipitation, while only two or three determinations, enough to check out the error, are necessary for the constant factors, altitude, slope, exposure, and surface. An account of the exact procedure employed in class study at Lincoln and Minnehaha will best serve to illustrate the use of this method. The series of stations chosen at Lincoln were primarily within a single formation, for the purpose of determining the physical factor variation in different areas. One series was located in the prairie-grass formation (_Koelera-Andropogon-psilium_), and consisted of the following stations: (1) low prairie, (2) crest of ridge I, (3) northeast slope of ridge I, (4) grassy ravine, (5) southwest slope of ridge II, (6) bare crest of ridge II, (7) thicket ravine. The other series was established in the bur-oak-hickory forest (_Quercus-Hicoria-hylium_) at the following stations: (1) thicket, (2) woodland, (3) knoll in forest, (4) depression in forest, (5) level forest floor, (6) nettle thicket, (7) brook bank. At Minnehaha the series was primarily one of different formations: (1) the pine formation (_Pinus-xerohylium_), (2) the gravel slide formation (_Pseudocymopterus-Mentzelia-chalicium_), (3) east slope of spruce forest (_Picea-Pseudotsuga-hylium_), (4) ridge in the spruce forest, (5) north slope of spruce forest, (6) brook bank in forest, (7) the thicket formation (_Quercus-Cercocarpus-lochmodium_), (8) the aspen formation (_Populus-hylium_). When permanent or temporary quadrats are established, they are ordinarily used as regular stations, since this enables one to refer the physical factor readings to a few definite individual plants, as well as to the entire formation. The transects in figure 29 illustrate two of the above series of stations. =133. Time of readings.= The frequency of simple readings and the times at which they are made must be regulated largely by opportunity and convenience. In addition to making readings once or twice a week throughout the season, the series should be read at least once every day for a representative week or two. It is also very desirable to have a series for each hour of a typical day, or of two days, one of which is clear, the other cloudy. When a single daily reading is made, it should be taken at or as near meridian as possible. The usual series is the one obtained by simultaneous observations at the same level in different stations. An important series is also secured by simultaneous readings at the various levels of the same station, though it is not necessary to take this series frequently. [Illustration: Fig. 30. A denuded station in the aspen formation.]