Table 2.

Estimates of parameters of the demographic model applied to Cross River and western lowland gorillas

	Prior distribution	min	max	Posterior mode	HDI 50b	HDI 90b	HDI 95b
2Nm	Loguniform	1	15.85	9.55	[4.57, 13.8]	[1.58, 15.84]	[1.32, 15.85]
Ncrossriver_old/Ncrossriver_now)	Loguniform	1	100	61.7	[33.1, 93.3]	[10, 100]	[4.2, 100]
Ncrossriver_now	N(200, 100) a	68	300	271	[223, 292]	[146, 300]	[122, 300]
Nancestral	Uniform	500	25,000	2,547	[1,383, 4,032]	[500, 6,681]	[500, 7,684]
Nwestern	N(24,000, 5000) a	10,000	30,000	22,376	[18,765, 25,319]	[14,217, 28,930]	[12,879, 29,532]
Tdivergence	Loguniform	10	3,162	891	[269, 1,738]	[60.3, 3,090]	[38, 3.162]
Tbottleneck	Loguniform	10	316	16	[33.1, 93.3]	[10, 97.7]	[10, 141]
Tmigration	Loguniform	10	3,162	21	[11.7, 60.3]	[10, 446.7]	[10, 812]

N_{crossriver_now}, N_ancestral, N_western represent the effective population sizes of Cross River, ancestral and western lowland gorillas respectively. Timings in generations were estimated on the log₁₀ scale and indicate the divergence (T_divergence), the onset of the bottleneck (T_bottlneck) and the cessation of migration (T_migration). The number of diploid individuals exchanged between the populations was also estimated on the log₁₀ scale as 2 Nm. For parameters estimated on the log₁₀ scale we chose uniform priors on the same scale.

^a Corresponds to a normal distribution of the form N(μ, σ) truncated at [min, max]

^b The high posterior density interval HDI is chosen as the smallest continuous interval spanning 50% of the posterior surface. The other HDI are chosen accordingly.