Table 1.

Possible outcomes (rows, columns) of a cross-infection experiment where parasite performance is assayed, both, on coevolved host populations, and on control host populations paired for the same original host lines.

			Control host populations
	Outcomes:	Mortality in matching combinations > non-matching combinations	Mortality in matching combinations < non-matching combinations	No difference between matching and non-matching combinations
	Mortality in matching combinations > non-matching combinations	1: Parasite adaptation	2: Parasite maladaptation < host maladaptation	3: Host maladaptation
Coevolved host populations	Mortality in matching combinations < non-matching combinations	4: Parasite adaptation < host adaptation	5: Parasite maladaptation	6: Host adaptation
	No difference between matching and non-matching combinations	7: Parasite adaptation = host adaptation	8: Parasite maladaptation = host maladaptation	9: No adaptation

The numbered entries refer to the interpretation for each possible combination of outcomes ^a. A matching combination refers to hosts and parasite being from the same experimental coevolution replicate; non-matching otherwise.

^a Differentiating between parasite adaptation and host maladaptation can be facilitated if parasite fitness is assayed on replicate host populations of identical genetic background as their coevolving antagonists. In this table we have used host mortality as a measure of both parasite and host performance, as these traits are very closely linked to fitness in both antagonists in our study system [39,41]. For other study systems, relevant traits may differ. This experimental approach has, to our knowledge, never been used, but could most readily be performed in a laboratory setting.