The same procedure was used for Rhus glabra, with the following modifications.
- A second method was used to estimate size of the soil seed bank of Rhus glabra using four additional soil samples (20 cm x 20 cm x 5 cm deep) collected as described above from each of the three clumps of this species.
For Rhus glabra, 24 seed traps (30 x 32 cm) made with the same wire netting mentioned above were set up under standing plants on 20 August 1994.
After one calendar year only 2 [+ or -] 1% of the Rhus glabra seeds in the unheated greenhouse had germinated [ILLUSTRATION FOR FIGURE 2B OMITTED].
- After 4 y in the unheated greenhouse, 900 [+ or -] 282 seedlings of Rhus glabra [m.sup.-2] had emerged.
Input to the soil seed bank in Rhus glabra was much greater than that in R.
That many more viable seeds of Rhus glabra remained impermeable (4525 [m.sup.-2]) than had germinated (900 [m.sup.-2]) after 4 y in greenhouse soil (Table 2) agrees with conclusions by other researchers that size of soil seed bank based on the direct seed-counting method greatly exceeds that determined via the seedling-counting approach (Olmsted and Curtis, 1947; Hurka and Haase, 1982; Gross, 1990; Brown, 1992).
For example, smooth sumac (Rhus glabra
), previously reported to be a major component in the diet of golden mice (Goodpaster and Hoffmeister, 1954; Linzey, 1968; Linzey and Packard, 1977), was of low food quality because of high levels of gallotannins when dietary experiments were conducted (Jewell et al., 1991).