Color plumage polymorphism and predator mimicry in brood parasites

Table 1 Great reed warbler responses to experimental dummy dyads (see Methods and Figure 1 )

Experiment	N	Preference	Continuous response		Categorical response
		(%)	Z	P	Z	P
Specific recognition
Gray cuckoo–Dove	18	100	85.5	<0.0001	45.5	0.0002
Rufous cuckoo–Dove	16	94	66.0	<0.0001	7.5	0.06
Kestrel–Dove	17	88	67.5	0.0005	7.5	0.06
Sparrowhawk–Dove	14	93	46.5	0.0016	5.0	0.13
Two morphs comparison
Gray cuckoo–Rufous cuckoo	18	72	48.0	0.035	0.0	1.00
Mimicry
Gray cuckoo–Sparrowhawk	23	96	135.0	<0.0001	7.5	0.06
Rufous cuckoo–Kestrel	20	70	35.0	0.20	0.0	1.00

Dummies within dyads that were mobbed more than their paired dummy are in bold. “Preference” is the percentage of warbler pairs that more strongly responded to the more attacked dummy (in bold) than to the paired dummy. The “Specific recognition” set of experiments asked “Do warblers recognize dangerous enemies near the nest specifically?” by comparing responses to cuckoos/predators with responses to harmless control turtle dove. The “Mimicry” set of experiments asked “Do gray cuckoos mimic sparrowhawks and do rufous cuckoos mimic kestrels?” N = number of host pairs. Responses were measured either as number of contact attacks (Continuous response) or re-coded as presence vs. absence of attacks (Categorical response). See Discussion for rationale behind and implications of categorical re-coding. Differences tested with Wilcoxon sign-rank tests.

ISSN: 1742-9994