The crucial role of earthworms in stabilizing soil organic matter and thereby stimulating carbon sequestration is widely acknowledged in the literature. However, there is a separate body of literature reporting elevated emissions of the greenhouse gas (GHG) nitrous oxide (N₂O) from earthworm-worked soil, earthworm casts, and earthworms themselves. As these last results are mainly derived from natural ecosystems, we quantified the effects of earthworms on N₂O and carbon dioxide (CO₂) emissions from a 3-year old pasture on a sandy soil in the Netherlands. In a first experiment, the effect of an anecic species (Aporrectodea longa) on emissions after grassland renewal was quantified for three different moisture contents. None of the moisture treatments showed a significant effect of A. longa on CO₂ emissions. At the highest soil moisture content (70% Water-filled pore space) presence of A. longa resulted in significantly higher N₂O emissions during the first two weeks, representing a difference of approximately 450 g N₂O-N ha^-1. However, after three weeks patterns reversed and N₂O emissions became significantly lower in the presence of A. longa for the remainder of the 2 month experiment. Potential denitrification and N₂O emission at the end of the experiment showed significant differences with respect to earthworm presence. In a second experiment, both the effect of an additional epigeic earthworm (Lumbricus rubellus) and of soil compaction were quantified, showing intricate interactions between earthworm species, compaction and GHG emissions. We conclude that the effects of earthworms on net GHG emissions from soils are not always exclusively beneficial and need to be studied further. Finally, we discuss the role of earthworms in the often-observed trade-off between C sequestration and N₂O emission after introduction of minimum tillage.