Author: Bryan P. White
Original Publication: 10/04/2012
Edited for grammar/syntax: 07/31/2019
It has been proposed that given enough time, even when the probability of dispersal across geographic barriers is very low, dispersal will eventually happen – a so-called ‘sweepstakes’ event. The Isthmus of Panama in Central America is one such geographic barrier that may lend itself to these so-called sweepstakes events, particularly for a marine species to disperse from one side of the isthmus to the other. The authors here hypothesized that some marine species might be more likely than others to win these dispersal sweepstakes, perhaps owing to differences in their natural history. The authors chose two, abundant, geminate, marine snails; the Pacific and Atlantic horn snails, which are geminate species in the genus Cerithidea. Preliminary evidence suggested that these snails have dispersed across the Isthmus land barrier, but in this study, those data were more broadly expanded on and more rigorously tested.
Samples were collected from 29 populations of both species across the coasts of North and Central America and sequenced every individual collected for their COI gene. Once COI-based clades were established, sub-sampling within each clade was done so that a more rigorous molecular clock phylogenetic analysis could be done. For these more rigorous samples, the 16S and 12S ribosomal genes were sampled, as well as the Cytb, NADH, and ND6. Nuclear markers were not sampled as they showed too much conservation to represent such a recent event as dispersal.
Phylogenetic analysis was conducted first using MODELTEST to determine the best nucleotide model for the data, which was determined to be GTR + I + G. They first used a strict molecular clock tree calibrated to 3.1 – 2.8 mya, the final closure of the isthmus, but the likelihood ratio test of the molecular clock was rejected. Next, they used a relaxed molecular clock model using BEAST, and the best nucleotide model for each gene was selected using MODELTEST, and the analysis was partitioned among those genes.
Tests of monophyly of each species were conducted using CONSEL, and biogeographic analysis was conducted using Arlequin.
A large amount of genetic variation was found in the COI analysis. There were two major clades (Clade A and Clade B), each of which also had several sub-clades. These sub-clades were distributed on both sides of Central America, which suggests there has been some recent gene flow across those barriers. Clade A had two very large clades on the Pacific, with only one sub-clade on the Atlantic. The opposite was found for Clade B, which had two very large sub-clades on the Atlantic and only one small sub-clade on the Pacific. This suggests that the Pacific clade arose and diverged in the Pacific, then crossed into the Atlantic more recently, and the opposite for the Atlantic clade.
Dispersal from the Pacific to Atlantic occurred around 750,000 years ago, and that dispersal from the Atlantic occurred 72,000 years ago. It is unlikely that the snails dispersed through some other marine channel, or through human interaction, as the isthmus was geologically closed until 1914. The authors suggest this dispersal is likely due to migrating shorebirds, which cross the isthmus regularly, and that snails can survive this dispersal even when ingested, and can later be regurgitated by the birds. Haplotype networks suggest that the dispersal of both clades actually occurred over the Isthmus of Tehuantepec, instead of the Isthmus of Panama, despite being larger. The presence of mitochondrial introgression is possible here, explaining some of the distribution of haplotypes.
I liked the methods of this paper. They were very straight forward, had a clear hypothesis, were easy to understand, and the data clearly supported their conclusions. This is both a good example of how older hypotheses such as low-probability dispersion events are still generating novel discoveries today, even though they have been suggested for over 200 years. This is also a great example of how investigators have streamlined the process, from data collection to conclusions, so that their methods can be easily applied to other study organisms or reproduced and replicated.
Miura, O., Torchin, M.E., Bermingham, E., Jacobs, D.K. and Hechinger, R.F., 2011. Flying shells: historical dispersal of marine snails across Central America. Proceedings of the Royal Society B: Biological Sciences, 279(1731), pp.1061-1067.