Barriers to dispersal

Barriers to dispersal may be relatively easy to recognize on land, for example water bodies and mountain ranges limit the distribution of many ground-dwelling species, and dry land is a barrier for many riverine species. In contrast, robust physical barriers are relatively uncommon in the ocean, and therefore oysters, clams, starfish, sea urchins and other species whose planktonic larvae can remain free-living for days, weeks or even months are theoretically capable of being transported across vast expanses of open ocean. Following on from this, it is intuitively appealing that long-lived larvae should have a higher potential for dispersal than short-lived larvae, and a number of comparative studies have supported this prediction. One example of this is sea urchins in the genus Heliocidaris. H. tuberculata has a planktonic larval stage of several weeks, and showed very little genetic differentiation between populations separated by 1000 km of ocean, whereas H. erythrogramma has a planktonic larval stage of only 3-4 days and showed much higher levels of genetic differentiation across a similar spatial scale (McMillan, Raff and Palumbi, 1992). This trend was summarized further by a study in which data from 19 species of benthic marine invertebrates revealed a significant correlation between the duration of the larval stage and the distances across which individuals dispersed (Figure 4.6).

It is tempting to conclude from data such as those summarised in Figure 4.6 that most, if not all, species with long larval periods will disperse widely on ocean currents, but although this is sometimes true, an increasing number of molecular studies have shown that trans-oceanic dispersal patterns are actually much more

t5 IS

600 500 400 300 200 100

600 500 400 300 200 100

Planktonic larval duration (days)

Figure 4.6 Comparison between the length of the planktonic larval stage and the mean dispersal distance (inferred from genetic differentiation) estimated for 19 species of marine benthic invertebrates. These data show that species with relatively long larval durations tend to disperse over relatively broad distances. Data from Siegel et al. (2003) and references therein

Planktonic larval duration (days)

Figure 4.6 Comparison between the length of the planktonic larval stage and the mean dispersal distance (inferred from genetic differentiation) estimated for 19 species of marine benthic invertebrates. These data show that species with relatively long larval durations tend to disperse over relatively broad distances. Data from Siegel et al. (2003) and references therein

complicated than this. For one thing, several taxonomic groups show little or no correlation between larval duration and gene flow, such as coral species in the Great Barrier Reef (GBR), which can be divided into species that brood their larvae (brooders) and species that broadcast their larvae into the water (broadcast spawners). Since the larvae of brooding corals typically begin to settle within 12 days of release, whereas the larvae of broadcast spawning corals usually drift for 5-7 days before settling, the latter may be expected to show relatively high levels of gene flow and accompanying low levels of genetic differentiation among populations. This, however, was disproved by one study that found similar levels of genetic differentiation and gene flow among populations of both brooders and broadcast spawners (Ayre and Hughes, 2000; Table 4.6).

Table 4.6 FST values calculated between different reefs within the Great Barrier Reef for nine species of coral. Four of the species are broadcast spawners and five are brooders. Note that there is considerable overlap in the FST values of brooders and spawners. Adapted from Ayre and Hughes (2000)

Species Mean FST Mean Nem

Brooders

Table 4.6 FST values calculated between different reefs within the Great Barrier Reef for nine species of coral. Four of the species are broadcast spawners and five are brooders. Note that there is considerable overlap in the FST values of brooders and spawners. Adapted from Ayre and Hughes (2000)

Species Mean FST Mean Nem

Brooders

Pocillopora damicornis

0.01

25

Acropora palifera

0.02

12

A. cuneata

0.05

5

Stylophora pistillata

0.09

3

Seriatopora hystrix

0.15

1

Broadcast spawners

Acropora millepora

0.01

25

A. valida

0.02

12

A. cytherea

0.03

8

A. hyacinthus

0.05

5

Other molecular studies of marine invertebrates have shown surprisingly low levels of gene flow between populations of species that have a high potential for dispersal. These seemingly paradoxical results sometimes have been attributed to previously unidentified physical barriers. Barriers to dispersal may be as simple as an expanse of open ocean. Five common and widespread coral species were sampled from several reefs within the GBR and also from Lord Howe Island. Within the GBR, gene flow was moderate to high across a distance of 1700 km. However, although Lord Howe Island is only 700 km further south, a lack of gene flow showed that dispersal there from the GBR was essentially non-existent (Ayre and Hughes, 2004). The authors of this study concluded that movements between the connected reefs of the GBR occurred in a stepping-stone manner, whereas the expanse of ocean between the GBR and Lord Howe Island was a barrier to dispersal. Similarly, the mantis shrimp Haptosquilla pulchella disperses across regions of semicontiguous coastlines that span thousands of kilometres, but populations separated by only 300 km across open ocean may experience very low levels of gene flow (Barber et al., 2002).

Barriers can also be formed by currents such as upwellings, which occur when winds move surface water offshore, causing deeper water to flow in the opposite direction. The cold-water upwelling close to the tip of South Africa has provided a barrier to long-term gene flow in a number of marine species, including shallow water sea urchins in the genus Tripneustes (Lessios, Kane and Robertson, 2003). There are also many large-scale ocean currents that influence dispersal, such as the Almería-Oran oceanographic front, which is a well-defined hydrographic boundary caused by a strong current running between Almería in Spain and Oran in Algeria. This has created a barrier to gene flow at the junction between the Atlantic Ocean and the Mediterranean Sea for a number of species that disperse as planktonic larvae, including the mussel Mytilus galloprovincialis (Ladoukakis et al., 2002) and the crustacean Meganyctiphanes norvegica (Zane et al., 2000). Barriers to dispersal may also be created by changing environments, for example the Rio de la Plata estuary in Uruguay hinders the dispersal of pelagic larvae of the crab Armases rubripes because the salinity and temperature are suboptimal for larval development and survival (Luppi, Spivak and Bas, 2003).

Tracking the dispersal of small species in such an expansive area as the open ocean was extremely difficult before the advent of molecular markers. It is becoming clear that, even if they are not readily apparent to the human eye, barriers will influence gene flow across virtually every type of habitat. In the next section we will return to terrestrial species to look at how reproductive ecology can influence patterns of dispersal and gene flow.

Was this article helpful?

0 0

Post a comment