Do pre- and post-copulatory sexually selected traits covary in large herbivores?

Table 3 Phylogenetically corrected models testing the relationship between weapon length and testes mass, sperm dimensions and mating tactics

Dependent variable	Independent variables	beta ± SE	t	P	N	df	r	λ
Weapon length	Body mass	0.44 ± 0.09	4.70	< 0.001	45	41	0.59	0.27^0.64/0.11
	Testes mass	−0.04 ± 0.12	−0.31	0.76			−0.05
	Tactic	−0.74 ± 0.60	−1.22	0.23			−0.19
	Testes mass × Tactic	0.11 ± 0.14	0.74	0.46			0.12
Weapon length	Body mass	0.56 ± 0.06	8.40	< 0.001	54	50	0.77	1.00^{< 0.001/0.81}
	Sperm length	−1.92 ± 1.05	−1.83	0.07			0.25
	Tactic	−5.12 ± 4.74	−1.08	0.28			0.15
	Sperm length × Tactic	1.20 ± 1.16	1.04	0.30			0.15
Weapon length	Body mass	0.53 ± 0.06	8.53	< 0.001	54	50	0.77	0.99^{< 0.001/0.38}
	Sperm head length	0.90 ± 0.76	1.18	0.24			0.16
	Tactic	2.14 ± 1.69	1.26	0.21			0.18
	Sperm head length × Tactic	−1.11 ± 0.81	−1.37	0.18			−0.19
Weapon length	Body mass	0.52 ± 0.07	7.50	< 0.001	53	49	0.73	0.98^{< 0.001/0.36}
	Sperm midpiece length	−0.26 ± 0.46	−0.56	0.58			0.08
	Tactic	−0.59 ± 1.46	−0.41	0.69			−0.06
	Sperm midpiece length × Tactic	0.16 ± 0.58	0.28	0.78			0.04
Weapon length	Body mass	0.49 ± 0.07	7.26	< 0.001	47	43	0.74	0.98^{< 0.001/0.20}
	Sperm midpiece volume	0.08 ± 0.17	0.45	0.65			0.07
	Tactic	−0.07 ± 0.23	−0.28	0.78			−0.04
	Sperm midpiece volume × Tactic	−0.06 ± 0.19	−0.34	0.74			−0.05
Weapon length	Body mass	0.55 ± 0.06	8.91	< 0.001	54	50	0.78	1.00^< 0.001/1
	Sperm tail length	−1.44 ± 0.75	−1.92	0.06			−0.26
	Tactic	−3.06 ± 3.14	−0.98	0.33			−0.14
	Sperm tail length × tactic	0.78 ± 0.85	0.91	0.37			0.13

Female defence species is coded as 0 and territory defence is coded as 1. Models are tested with Bininda-Emonds et al. [31] phylogeny across ungulates species (both Bovidae and Cervidae pooled). r represents effect size values and λ represents the index of phylogenetic covariance (see Methods section). The superscripts following the λ value indicate p-value of likelihood ratio tests against models with λ = 0 (first position) and λ = 1 (second position). All variables were log transformed.

ISSN: 2730-7182