Metasiro (Davis, 1933)

Clouse, Ronald M., Sharma, Prashant P., Stuart, Jessie C., Davis, Lloyd R., Giribet, Gonzalo, Boyer, Sarah L. & Wheeler, Ward C., 2016, Phylogeography of the harvestman genus Metasiro (Arthropoda, Arachnida, Opiliones) reveals a potential solution to the Pangean paradox, Organisms Diversity & Evolution (New York, N. Y.) 16 (1), pp. 167-184 : 179-180

Metasiro dispersal and population histories

Our results are consistent with a scenario in which there is normally very little gene flow between populations, but rare long-distance dispersal events do occur. This is seen most clearly in M. americanus in Florida, whose history appears to include both (1) periods which may last many thousands of years when populations remain genetically isolated at scales of only a few dozen meters; and (2) episodes when individuals are passively carried across distances that, for a mite harvestman, are large. We are not aware of any other non-sessile, non-cloning, non-selfing animal where gene flow is restricted at such short distances, absent of cataclysmic events, and this is consistent with studies showing, with few exceptions, cyphophthalmid evolution and geologic history matching each other closely. The only possible similarity is with the effect of settlement of larvae in some marine invertebrates, showing a drastic decrease in genetic relatedness over the first 100 cm of distance ( Calderon et al. 2007), but this is probably due to kin recognition effects.

Within this broad scenario, Metasiro populations do have different histories, as evidenced by their genetic diversity, haplotype relationships, and tests of population stability. In Florida Caverns State Park, we discovered a haplotype (no. 103) that is quite different from and distantly related to the other haplotypes in that locality. We also observed this in the Torreya populations, which include one haplotype that is most closely related to those in Upper Sweetwater Creek. Haplotype 101 in the Florida Caverns also gave evidence of past gene flow. This haplotype was collected at both the hillside log and the upland glade. However, it is not indicative of ongoing gene flow between the hillside and the glade, for no other haplotypes are shared between the two localities, not even the common haplotypes 106 and 103. The Florida Caverns populations are also distinctly less diverse than those along the Apalachicola River, and the common flooding in the Florida Caverns area (large areas had been inundated only days before our collections) may promote frequent population bottlenecks. Consistent with this, our estimate of the effective population size for the upland glade (30,000 –90,000) is one to two orders of magnitude lower than other sites. This process would also be likely to happen in the lowland forests of the Savannah River delta, and indeed M. savannahensis shows clear evidence of a population bottleneck.

The habitat of Metasiro could be the cause of more than just bottlenecks, for flooding could move individuals in floating mats of debris. New populations could arise from mixtures of local survivors and individuals relocated from distant areas, creating the strange effect of unrelated haplotypes in the same population. Metasiro live deep in thick layers of fine detritus (the B duff^ layer) where air pockets likely provide refuge during brief floods and could aid in keeping debris afloat. An interesting line of inquiry would be to test experimentally whether flood waters have floating debris capable of carrying individual cyphophthalmids to new localities.

Flooding along the Apalachicola River can be caused by local storms but also by high water coming down the Chattahoochee River from the Southern Appalachians. The southern Appalachians and especially northern Georgia can experience heavy rains that increase the flow rate in local rivers in excess of 1000-fold in a few days ( McCallum and Gotvald 2010), bringing debris to the lowlands. Moreover, the other widespread mite harvestman for which we have phylogeographic data, Aoraki denticulata , lives at a variety of elevations among the rivers of the Tasman and Nelson regions of New Zealand. This is also an area prone to flooding ( NIWA 2010), which is generally one of New Zealand’ s most prominent hazards ( Smart and McKerchar 2010). It was suggested that past geologic events isolated A. denticulata populations, causing the evolution of highly divergent COI haplotypes, and now the populations have merged and experience a small amount of gene flow between them ( Boyer et al. 2007a). What we propose here for M. americanus is the opposite: isolated populations normally experience no gene flow among them, but gene flow happens on a rapid and massive scale among various subsets of populations during historical floods. We also notice that M. sassafrasensis lives at the head of both the Apalachicola and Savannah River watersheds. The three odd species of Metasiro are not as disjunct as they first appear, and these watersheds along with major floods could have been the conduit by which Metasiro expanded after ocean incursions. This demographic model may apply to A. denticulata as well, which has been reconstructed to have radiated from the mountains to the coast and which shows signs of haplotype mixture in local populations ( Fernández and Giribet 2014).

Geologic history of the Southeastern United States and dating Metasiro

The dates we recovered for the history of M. americanus using a date-range calibration of Sternophthalmi from Giribet et al. (2012) in PATHd8 tightly match historical events in the region. Contrary to our expectations, M. sassafrasensis in the Southern Appalachians is sister to the other two species and appears to be the oldest of the three, and the variety of dates we recovered from tests in PATHd8 are consistent with a scenario in which Metasiro existed as a relict in the Southern Appalachians for many millions of years between the breakup of Pangea and the retreat of the Atlantic and Suwannee Strait. This relictual nature of Metasiro is reminiscent of the cyphophthalmid family Troglosironidae on New Caledonia ( Sharma and Giribet 2009), which is also hypothesized to have expanded quickly once land exposure increased. The most recent event in the diversification of Metasiro of interest here, and the one most removed from user-input calibration points, was the start of diversification of M. americanus in the western panhandle of Florida. We would expect this to match the retreat of the Gulf Trough 14 Ma, and we recovered mean dates of 12.9–15.9 Ma from different dating tests. Thus, fidelity between the dated phylogeny of harvestmen and geologic history appears to have remained remarkably unchanged since the Paleozoic. As independent tests of this hypothesis, future endeavors should add Metasiro specimens from Georgia, with the expectation that divergence of the Georgia lineages accords with the timing of exposure of the inhabited terranes.