Table 1.

Generalised linear models for a) amplification success and b) polymorphism of conserved microsatellite loci

a) Amplification success1						b) Polymorphism2
	Error	Residual	B	t	P		Error	Residual	B	t	P
	df *	deviance*					df *	deviance*
Maximal model	39	57.29				Maximal model	12	24.97
- Interactions	49	62.32				- Microsatellite retained	13	24.97
- Repeat length	50	62.32				- Primer type5	14	25.46
- Unique/multiple hits	51	62.70				- Ho in source species6	17	26.91
- Microsatellite motif3	52	63.55				- No. of mismatches7	18	27.60
- No of mismatches4	53	67.65				- E-value	19	30.15
Minimal adequate model						Minimal adequate model
E-value	53	67.65	-0.02	-2.58	0.013	Repeat length	19	30.15	0.02	2.22	0.039
						Microsatellite motif3	19	30.15	-1.00	-2.15	0.045
						Microsatellite interrupted8	19	30.15	-1.02	-2.09	0.051

*model descriptives are given for models after removal of explanatory variables

¹based on tests in three charadriiform species: Kentish plover Charadrius alexandrinus, whiskered auklet Aethia pygmaea and ruff Philomachus pugnax

²based on proportion of species in which a microsatellite locus was found to be polymorphic when tested in four individuals

³dinucleotide or tetranucleotide repeat motif of microsatellite

⁴total number of mismatches between chicken and charadriiform sequence for both primer sequences combined

⁵Gga-consensus primer or standard primer design

⁶observed heterozygosity for source species given in original publication of the microsatellite sequence

⁷number of mismatches in primer sequence between charadriiform sequence and chicken homologue

⁸microsatellite sequence interrupted by non-motif base insertions (imperfect) or not (perfect)