The members of the Pea secundo complex were studied udng transphtnt experiments, morphological studies of population samples, and various nmnericnl taxonomic techniques including principal components analysis and discrhninant analysis. The complex is shown to comprise 2 species: Poa curtifohz, a serpentine endemic from central Washington, and P. secundu, a widespread polymorphic rangegrass. Other forms may be recognizable locally, but do not represent separate evolutionary lines. If range managers need names for these local forms, the names should be informal English names rather than Latin binomials. Hitchcock (1950), Hitchcock et al. (1969) and Cronquist et al. (1977), give keys to the bluegrasses, but anyone who has tried to identify the grasses has found it very difficult: either the keys need improvement or the “species” are not really distinct. In the Poa secundu Presl group, there are about 40 published names; Hitchcock (1950) recognized 8 species, Hitchcock et al. (1969) recognized 11, and Marsh (1952) only 1. This study was designed to analyze the variation among the plants in the group to determine how many species could be recognized, their distinguishing characters, and the appropriate scientific names (see also Kellogg 1985). This paper summarizes the results most relevant to range scientists and managers, and then discusses the relationship of taxonomy and evolutionary biology to applied biology and management. Such a study should interest range managers and biologists for 2 reasons: First, a taxonomic study often results in name changes, important only because a standard vocabulary makes communication possible among range managers. Second, and more important, is a greater understanding of the biology of the plants. Plants and Names Applied Poa secundu (= P. sundbergii Vasey), a common bunchgrass, is most familiar as a small (30 to 50 cm) caespitose plant that becomes active in early spring. After flowering it turns reddish and becomes dormant for the rest of the summer. Several other forms (P. amplu Merrill, P. cunbyi(Scribner) Piper, P. curtifoliu Scribner, P. grucillima Vasey, P. incurvu Scribner & Williams, P. juncifoliu Scribner, P. nevudensis Vasey, and P. scubrellu (Thurber) Bentham) are closely related to P. secundu, and are distinguished by minor morphological characters. Some of these “species” are of interest to range managers because they are better forage grasses than typical P. secundu. For instance, P. umplu, P. cunbyi, P. juncifoliu, P. nevudensis, and P. scubrellu all tend to have longer basal leaves, and P. scubrellu and P. umplu continue growth longer into the summer. The group is distributed over most of western North America, with disjunct populations in the GaspC peninsula of Quebec and others in Chile. The plants occur on both moist and dry sites, from areas such as the Mojave Desert to high alpine areas of the Sierras and Rocky Mountains. The group has been called the Pou sundbergii complex, but Arnow (1981) has shown that it includes some populations in Chile, the name of which, Pou secundu, is older and takes precedence. Although the members of the complex are commonly distinguished from other bluegrasses by the lack of a keel on the lemma, I have found this an unreliable character. It is easier to distinguish Author is a postdoctoral fellow at the Harvard University Herbaria, 22 Divinity Avenue, Cambridge, Massachusetts 02138. Thanks to D.M. Henderson, under whose direction this study was started, and to C.E. Wood, Jr., under whose direction it was completed. This manuscript was greatly improved by theeditorial helpof SM. Rossi,and bythecommentsof twoanonymous reviewers. Manuscript accepted April 18, 1985. 516 the Pou secundu group by both its longer rachilla internodes (0.6 to 1.9 mm vs. <0.5 mm for most other members of the genus) and its longer spikelets (7 to 10.5 mm vs. < 6 mm for many, but by no means all, other species). The anthers are also longer than in many other bluegrasses (1 to 4.2 mm vs. <I mm). All members of the group are bunchgrasses with intravaginal branching; most plants have narrow panicles. The perfect flowers, although frequently pollen-sterile, are often apomictic. Long tangled cobwebby hairs, characteristic of such species as P. prutensis, are lacking.
Read full abstract