Predictable patterns of trait mismatches between interacting plants and insects

Table 2 Summary of the models examined to explain variation in morphological patterns of trait matching between insect and plant species.

Phylogeny	Analysis	λ lambda	AIC	w _i	ΔAIC	Slope± SE	UCL:LCL	P	Df
	Conventional		1031.1	<0.0001	36.82	0.783 ± 0.023		<0.0001	135
Insect	Phylogenetic	0.797	988.6	>0.999		0.645 ± 0.028	0.700: 0.590	<0.0001
Plant	Phylogenetic	0.499	1025.4	<0.00001	42.52	0.765 ± 0.025	0.815: 0.715	<0.0001
	RMA					0.827 ± 0.022	0.784: 0.870	<0.0001

Each model was tested using either (1) an actual phylogenetic hypothesis for one taxon (plants or insects), assuming equal branch lengths (see Methods), or (2) a "star-shaped" phylogeny with all species equally related, equivalent to a conventional non-phylogenetic analysis or (3) using reduced major axis (RMA) regression without any phylogenetic adjustment. LCL: lower 95% confidence limit; UCL: upper 95% confidence limit. λ lambda is a measure of phylogenetic correlation (for details of how this is calculated see Halsey et al. 2006); AIC is Akaike's information criterion; w_i is the Akaike weight, the probability that the model is the correct one of those tested [67]. RMA confidence limits calculated by bootstrapping (1000 bootstraps) in RMA for JAVA v. 1.21 [63].

ISSN: 2730-7182