Table 1.

Biogeographic models analyzed in BioGeoBEARS. Each dispersal or vicariance scenario was tested using the six models available in BioGeoBEARS (DEC, DEC+J, DIVALIKE, DIVALIKE+J, BAYAREALIKE, BAYAREALIKE+J)

Time Period of Model	Geologic Events Reflected in Model	MODELS 1: Without GAARlandia | 2: With GAARlandia
		a: Dispersal probability decreases as distance increases	b: Distance does not affect dispersal probability	c: Probability of overwater dispersal is very low
A: 3 my – present	Complete closure of the Central American Seaway and uplift of the Isthmus of Panama [18].	A1a | A2a	A1b | A2b	A1c | A2c
B: 23 my – present	First collision between southern tip of Central America and northwestern South America [19].	B1a | B2a	B1b | B2b	B1c | B2c
C: 15 my – present	Significant land above sea level or complete closure of the Isthmus of Panama [19, 20].	C1a | C2a	C1b | C2b	C1c | C2c
D: 8 my – present	Shoaling of the Central American Seaway from 12 to 7 my [21]; land bridge nearly complete by 10 my [32]. * We chose to use this intermediate period to account for the gradual closure of the Central American Seaway and its uncertainty after 15 my and before 3 my.	D1a | D2a	D1b | D2b	D1c | D2c
E: 23–15 my AND 3 my – present	This model emulates dispersal possibilities during two periods: that of the first contact of South America and Central America, followed by the full closure of the Central American Seaway.	E1a | E2a	E1b | E2b	E1c | E2c
F: 15–8 my AND 3 my – present	This emulates dispersal possibilities during two periods: during a time when there may have been a significant amount of land, followed by the full closure of the Central American Seaway.	F1a | F2a	F1b | F2b	F1c | F2c
G: 8–5 my AND 3 my – present	This emulates dispersal possibilities during two periods: a time when there was shoaling of the Central American Seaway and an increase in land area of the Isthmus of Panama, followed by the full closure of the Central American Seaway.	G1a | G2a	G1b | G2b	G1c | G2c