Is sperm viability independent of ejaculate size in the house cricket (Acheta domesticus)?
Assessing sperm viability is a popular means of testing hypotheses related to ejaculate quality. This technique has produced interesting results; however, the sperm viability assay itself may kill sperm. This is a serious pitfall, as assay-related mortality could confound results and produce artificially low estimates of viability. To avoid spurious results, it has been recommended that investigators include sperm number in their viability analyses. Unfortunately, studies conducted to date on internal fertilizers have not included sperm counts in their analyses, so it is not possible to assess whether this factor can indeed produce artefactual results. In this paper, we use male house crickets ( Acheta domesticus (L., 1758)) to show that sperm viability is dependent on sperm number and exclusion of this factor from statistical analyses affects our interpretation of experimental treatment results. We show that mechanically rupturing a spermatophore significantly reduces sperm viability, but this does not appear to drive the nonindependent relationship between viability and number. Instead, our study shows that nonindependence is due to processes other than differential physical damage to the sperm during collection. We also show that allowing a spermatophore to evacuate its sperm without rupturing for 10 min maximizes both sperm number and viability.