An 11-year Ecological Study of a Rare Prairie Perennial (Asclepias meadii): Implications for Monitoring and Management.(Statistical Data Included)

W. DEAN KETTLE [1]

HELEN M. ALEXANDER [2,3]

GALEN L. PITTMAN [1]

ABSTRACT.--Mead's milkweed, Asclepias meadii, is a rare long-lived perennial of North American tallgrass prairies. Stems of this clonal species are spatially aggregated and, therefore, observing survivorship and flowering of "patches" of stems best approximates the fate of genetic individuals. Population size is likely to be underestimated because more than one genotype can sometimes occur in a patch. The number of patches detected at a site in Kansas has greatly increased over the last 11 y since marking locations of individual patches allowed detection of nonflowering stems in subsequent years. The A. meadii population at our site (managed by biennial dormant-season burning) often had more flowering ramets and produced more mature follicles in years with burning. High rainfall in the preceding year, in conjunction with burning, was associated with the highest follicle production. The difficulties in detection of plants at the site mean that counts of numbers of patches over an 11-y period cannot be used to assess whether the population is increasing, decreasing or remaining constant. Several factors indicate a positive outlook for the population: management (burning) enhances fruit production, patch survivorship is high and a likelihood that more patches exist than are counted. However, the low fruit production in most years at the site is a concern for the long-term viability of the population.

INTRODUCTION

Although grasses dominate the biomass of prairies, forbs are responsible for the high species diversity of prairies of central North America (Great Plains Flora Association, 1986). Many forb species in prairies are long-lived perennials (Hartnett and Keeler, 1995) with lifespan estimates, inferred from survivorship of established plants, ranging from decades to centuries (Weller, 1985; Keeler, 1991; Pleasants and Jurik, 1992). Such longevity has conservation implications since it means a plant population may persist at a site for many years, even if the population has a net negative growth rate due to low reproduction. Asclepias meadii Torr ex A. Gray, Mead's milkweed, is a rare long-lived perennial that may fit this scenario. Individual plants have been followed for at least 25 y and it has been postulated that some individuals may live for a century (Betz, 1989). Fruit production is often low (6.4% of flowering stems producing follicles, Betz, 1989). Despite its potential longevity, continued loss of prairi e habitat has led to the current restricted distribution of A. meadii centered in Kansas and western Missouri (Betz, 1989; Bowles et al., 1998) and its listing as a threatened species under the Endangered Species Act (United States Fish and Wildlife Service, 1988).

The prairie flora evolved under a fire regime (Collins and Gibson, 1990) and Asclepias meadii seems to respond favorably to burning (Bowles et al., 1998). However, the majority of the 150 known populations of A. meadii exist in prairies maintained as hay meadows, which are unburned sites that have been mowed for decades (Bowles et al., 1998). Because these hay meadows are summer-mowed, mature fruit (follicles) cannot form and thus plants persist but there is no sexual reproduction. Recent studies show that A. meadii populations in the few prairies managed by burning have greater genetic diversity than those from hay meadows (Tecic et al., 1998).

Assessment of the status of Asclepias meadii populations is difficult because of characteristics of both the species and its habitat. Asclepias meadii plants (=genets) consist of one to several stems (=ramets) that are produced from rhizomes; clonal spread occurs but is not extensive (patches of stems at our site generally do not exceed 1 m in diameter). Flowering stems of A. meadii are easily found, but monitoring is difficult because plants do not flower every year (Beta, 1989) and nonflowering stems are hard to detect in dense prairie vegetation. The long period before first flowering (3-7 y in cultivation and thought to be much longer in the field) (Beta, 1989; Bowles et al., 1998) and the difficulty in detecting seedlings makes assessment of recruitment difficult.

The overall goal of our 11-y study was to determine the abundance, distribution and reproductive output of Asclepias meadii in a native prairie site that is maintained by burning in northeastern Kansas. Specifically, we: (1) documented the number of stems and patches at our site, (2) evaluated whether yearly variability in stem numbers and size (flowering and nonflowering ramets) and follicle production was related to management (burning) and rainfall and (3) combined our ecological data with limited genetic information to determine if patches are equivalent to genetic individuals.

METHODS

Study organism.--Asclepias meadii is a perennial herb that grows from a shallow slender rhizome (Great Plains Flora Association, 1986). Mature plants may produce from one to several stems. Individual stems, which flower in late May--early June, usually produce a single terminal umbel with an average of 9-12 flowers (Beta, 1989; Bowles et al., 1998). Flowering within a population occurs over a 10-12 d period with individual flowers open for 5-6 d (Bowles et at, 1998). Pollinators include bumble bees (Bombus) and miner bees (Anthophora) (Beta et al., 1994). The plant appears to be self-incompatible, as is common among milkweed species (Wyatt and Broyles, 1994; Bowles et at, 1998). The fruit is a pair of follicles, but often only one develops (Great Plains Flora Association, 1986). In our discussion we have used the technical term "follicle" rather than the nontechnical term "pod" which is often applied. Follicles dehisce in late August-early September releasing wind-dispersed seeds. Ascelpias meadii was former ly distributed throughout the tallgrass prairie of the midwestern U.S.A., but habitat destruction has greatly reduced its abundance (Beta, 1989; Bowles et al., 1998).

Study site.--The 4.5-ha site is on the Kansas …