It is a pleasure to acknowledge the assistance and inspiration of many people, particularly A

Rapidly elongating Bromus tectorum and Taeniatherum asperum roots penetrated the soil ahead of Agropyron spicatum roots and used available moisture. In contrast, Agropyron desertorum roots penetrated the soil almost as rapidly as B. tectorum and T. asperum and remained in favorable moisture. These differences in root penetration resulted in lower leaf water potentials and poorer survival in A. desertorum. The results suggest that in areas where root growth occurs at low temperatures and where lands are infested with B. tectorum and T. asperum, seedlings of A. desertorum would be more successful than seedlings of A. spicatum.

Highlight: Roots of plants express phenology significant to readily invade and become established on disturbed sites. competitive relationships, as do aerial parts. The following devel-Favorable root phenology is one of the adaptive strategies opmental stages of root phenology are proposed: (1) germination, allowing this superior competitive ability. (Perhaps the same (2) initial root growth, (3) rapid extension of root-soil contact, (4) dormancy, and (5) death. Examples of root phenological develop-principles apply to annual and perennial forbs.) The following ment in seedlings of three grass species are given to demonstrate discussion relates primarily to competition between seedlings of &ects on competition. annual and perennial grasses. Range plants universally compete with each other for the limited resources (light, water, nutrients, soil, air) of their ecosystem, Lacking human or biological interference, competition is a major factor in the determination of composition in a given plant community. Plants have many adaptations (morphological, anatomical, physiological, and phenological) suiting them to a place in the ecosystem. Among these is a characteristic phenology or response to seasonal change. Timing of the developmental stages of aerial plant parts (vegetative, reproductive, and dormant) has significant survival value in competition situations; timing of phenological development in roots, though less obvious, similarly has important survival value. A recent literature search failed to disclose any reference to studies of toot phenology, or to the development of theories of the function of variations in root phenology in plant competition. The author has conducted a line of research over the years, aimed at discovering the controlling factors in competition between seedlings of important native and introduced perennial and annual grasses of intermountain west rangelands. This work has led to investigations of rooting habits, including phenology, which have significance in understanding plant competition. Published results of seleL_ed experiments are recounted here in the development of conclusions principle of plant competition. regarding root phenology as a Stands of perennial grasses have a natural competitive advantqe over annual grasses. It is not necessary for them to begin ,T;~o seed following each aormant period. However, once mature plants are removed, and reestablishment from seed required; perennials are at a disadvantage. Seedlings of annuals Author is chairman and professor,

Manual methods for estimating root lengths in washed samples are time consuming, tedious, and subjective. The objective of this research was to evaluate the use of an inexpensive commercial image analysis system for measuring the length of roots in samples washed from soil. The analyzer consists of a high‐resolution television camera and a comparator, which measures the intersections of the TV scan lines and roots in the view area. Mathematical corrections are derived for overlapping roots in the sample and for limitations in system resolution. Errors due to nonrandom orientation of roots were minimized by scanning in two directions. Methods for contrast and light control are discussed. On a 1000 cm simulated root sample, the uncorrected length measurement was < 50% of the actual length. Correcting for overlaps increased the measured length to around 70% of actual length. Overlap and resolution correction resulted in errors less than 5%. Coefficients of variation (CV) for repeated measurements on a single, undisturbed wheat (Triticum aestivum L.) root sample were around 0.5%, while errors resulting from nonrandom dispersion of samples in the tray were 2‐3%. The image analysis system therefore appears to capable of providing reliable root measurements.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology. Abstract. Plant H20 relations and soil moisture depletion and recharge were followed in a stand of Artemisia tridentata near Washtucna, Washington during 1973 and 1974. Precipitation during the 1972-73 recharge season was 14.5 cm, 11 cm below normal. The 1973-74 precipitation was 35.7 cm, or 10 cm above normal. The 2 yr were therefore ideal for comparing plant behavior on wet vs. dry years. Soil moisture was depleted to around -70 bars in 1973 and -60 bars in 1974 to depths of 2.5 m. Leaf H20 potentials were -10 bars in the spring and decreased to -50 to -60 bars in the summer of 1973. In 1974, summer leaf H20 potential was -30 bars. Osmotic potentials were around -20 bars in the spring of 1973 and throughout the spring and summer of 1974. In the summer of 1973 osmotic potentials dropped to -60 bars. Stomatal diffusive conductances of 0.5 cm/s (resistance of 2 s/cm) were common during the spring and summer of 1974 and the spring of 1973. Summer conductances in 1973 were .05 cm/s (resistance of 20 s/cm).A simple model is proposed which predicts evaporation and transpiration from daily precipitation and potential evapotranspiration. Model parameters include soil hydraulic properties and maximum transpiration rate as a function of available soil moisture. The agreement between model prediction and measured values is within the uncertainties imposed by the input data. The model predicts 6.5 and 8.2 cm of evaporation for 1973 and 1974 respectively. In 1973 this was half of the total precipitation received. In 1974 it was only one fourth of the total.