Earthworm presence and potential impact on soil seed banks in northeastern Ohio forests

Background/Question/Methods

The floras of many U.S. woodlands have been decimated due to overbrowsing by white-tailed deer.   Plant regeneration will depend on soil seed banks whose stores of seeds may be threatened due to the foraging activity of introduced earthworm species.  This study sought to establish the potential for earthworms to deplete soil seed banks in northeastern Ohio.  In late May and June of 2013, earthworm populations were surveyed by using a mustard extraction technique.  Extractions were done  in twenty one 0.25 m² plots located at 10 m intervals along 3 transects in the Hinckley Reservation of the Cleveland, Ohio Metroparks. Soil samples from each plot were tested for seed germination under growth chamber conditions.  Feeding trials also were conducted in a growth chamber to determine if Lumbricus terrestris or Lumbricus rubellus would consume seeds from local plants.  Each of 102 L. terrestris and 87 L. rubellus was provided with 50 seeds from one of the following understory plants: Arisaema triphyllum, Dicentra cucullaria, Geranium maculatum, Sanguinaria canadensis, Solidago flexicaulis, or Thalictrum dioicum.  The number of seeds ingested during an 18 hour trial minus those egested 48 hours after a trial ended was used as an estimate the number of digested seeds.

Results/Conclusions

The worm extractions produced 312 earthworms belonging to 4 genera: Allobophora (8%), Apporectodea (31%), Lumbricus (35%), and Octolasion (24%).  The number of worms per plot ranged from 0 to 34 and averaged 15.6.  The Lumbricus species were either L. terrestris or L. rubellus. During feeding trials, the mean number of seeds digested by L. terrestris ranged from 12.7 for  S. flexicaulis to 0 for A. triphyllum.  L. rubellus consumed significantly fewer seeds than L. terrestris (two factor ANOVA, F_1,177=16.5, p<.001),  but showed the same pattern across plant species with an maximum mean of 5.5 for S. flexicaulis and a minimum mean of 0 for A. triphyllum.  Seed size (length or diameter) was negatively correlated with the number of seeds digested (r = -0.365, 186 d.f., p<.001).  Very few germinations were recorded from the soil samples: 15 from June 2013 samples and only 4 from October 2013 samples.  The number of worms observed in our sample plots combined with their capacity to digest especially the smaller seeds is consistent with our germination results and suggests that earthworms are depleting soil seed banks.