Legumes May Be Symbiont-Limited during Old-Field Succession
Larson, Jennifer L., Siemann, Evan, The American Midland Naturalist
JENNIFER L. LARSON' AND EVAN SIEMANN2
ABSTRACT.-We employed a bioassay using soybean seeds and native prairie bush clover (Lespedeza capitata) seeds to demonstrate that legumes may be rare throughout secondary succession in nitrogen-poor grasslands due to a lack of suitable rhizobia and consequently lower growth rates.
In many systems (Vitousek and Howarth, 1991), including Cedar Creek Natural History Area (Tilman, 1984, 1987), nitrogen is the nutrient most limiting plant growth, and through succession soil nitrogen concentration increases (Gorham et al, 1979; Inouye et al., 1987). At low soil nitrogen concentrations, nitrogen-fixing legumes can be competitively superior to nonlegumes. However, studies at Cedar Creek have found that legumes are always rare and their abundance is unrelated to field age and soil nitrogen content (Inouye et al., 1987). Currently there is no clear explanation for their rarity, although herbivores (Ritchie and Tilman, 1995), low relative growth rates (Tilman, 1982) and higher requirements for other nutrients compared to nonleguminous plants (Tilman, 1982) may have an effect. They may also be at a competitive disadvantage due to a lack of compatible rhizobia that limit the formation of nodules.
Rhizobia differ in their ability to infect different species of plants (Bergey et al., 1984; Paul and Clark, 1989; Prescott et al., 1996) . After abandonment of a field previously planted with soybeans, there may initially be an abundance of rhizobia. Through time, the numbers of these rhizobia may decline if their corresponding symbiont is not present (Paul and Clark, 1989; Kucey and Hynes, 1989). Therefore, we hypothesized that the ability of soybeans to form nodules will decrease with time since last cultivation. Because rhizobia for native legumes may come from accumulation of rhizobia in the plant-root environment of colonizing legumes (Weaver et al., 1971), we hypothesized that the number of nodules on roots of native prairie legume seedlings may increase with time since abandonment. In order to estimate the nodulation potential of legumes in fields of different successional ages, we used a bioassay employing Lespedeza capitata (bush clover, a common native prairie legume at Cedar Creek) seeds and soybean seeds.
MATERIALS AND METHODS
This work was performed with soils from 20 grassland fields at Cedar Creek Natural History Area which is located approximately 50 km N of Minneapolis, Minn. One field was last planted with potatoes, three with corn, one with oats, eleven with soybeans, four with rye and the crop history of one field is unknown (Inouye et al., 1987). Field ages were 10, 19 (four fields), 21, 31, 35, 38, 39 (two fields), 44 (two fields), 45 (two fields), 46, 53 (three fields) and 60 yr (Inouye et al., 1987). All fields had sandy soils that belong to the same taxonomic group, a Typic Udipsamment (Grigal et al., 1974; Johnson et al., 1991).
In July 1996, five pairs of 2.5 cm X 30.5 cm soil cores were taken at 1.5 m intervals along an existing 40-m transect in each of the 20 fields (same as used by Inouye et al., 1987). The two soil samples within each pair were mixed and split into two samples, one for a Lespedeza seed and one for a soybean seed (five of each species for each field). In order to limit the bias of nodule growth results, the sample location was first surveyed for the presence of legumes within 1 m. If legumes were present in the desired collection location, the sample was taken from 1 m away. Because legumes were rare in these fields, this eliminated the bias from sampling at a legume but may have biased our rhizobia abundance estimates to be slightly too low.
Before planting, the Lespedeza and soybean seeds were surface sterilized to reduce rhizobia numbers from the seed surface, and scarified to increase the germination of the seeds using the sulfuric acid soak method (Somasegaran and Hoben, 1994). …