What Influences Gene Flow

There are both advantages and disadvantages to dispersal. Dispersing individuals may benefit by avoiding inbreeding, locating a new site with relatively few

Table 4.4 A summary of the attributes of different estimates of gene flow

Indirect (Nem)


Assignment tests


Population equilibrium, marker neutrality, island model

Number of populations feasibly studied

Dispersal or gene flow


Gene flow

Likelihood of detecting long-distance dispersal

Low if long-distance dipersal is rare

Dependence on Moderate sample size

Direction of dispersal identified?

Capture-mark-recapture, radio tracking: negligible. Parentage: individual-specific genetic profiles

Mark--recapture, radio tracking = dispersal. Parentage gene flow

Radio tracking: high. Mark-recapture, parentage: low

Radio tracking: low. Mark--recapture, parentage: high

Linkage equilibrium, Hardy-Weinberg equilibrium



Theoretically high, as long as migrants'

source populations are genotyped


Yes competitors, or escaping from pathogens, parasites or predators. On the other hand, they may be unable to locate a suitable new site or mate, or they may be preyed upon en route. From an evolutionary perspective, dispersal should occur only if the benefits outweigh the risks, and this cost--benefit analysis is part of the reason why the frequency, mode and distance of dispersal vary tremendously between species.

Dispersal patterns are particularly complex in plants, fungi and invertebrates because often they have multiple dispersal mechanisms. The complexity of this is illustrated by the freshwater bryozoan Cristatella mucedo, which is a sessile, colonial, benthic species that can reproduce vegetatively and also through the production of larvae and seed-like propagules (statoblasts). The adult colonial stage can disperse only very slowly by creeping along substrate, and an individual colony is unlikely to move more than a few centimetres during the growing season. Sperm are broadcast into the water column, providing a means for local dispersal of the paternal genome. Similarly, larvae are free-swimming, but even if they are swept along in a current both sperm and larvae are short-lived and are unlikely to

Figure 4.4 Scanning electron micrograph of statoblasts (approximately 0.8 mm in diameter) from the bryozoan Cristatella mucedo that have become attached to a feather with the aid of small hooks. This is an important mechanism for the dispersal of this species between lakes and ponds. Photograph provided by Beth Okamura and reproduced with permission.

Figure 4.4 Scanning electron micrograph of statoblasts (approximately 0.8 mm in diameter) from the bryozoan Cristatella mucedo that have become attached to a feather with the aid of small hooks. This is an important mechanism for the dispersal of this species between lakes and ponds. Photograph provided by Beth Okamura and reproduced with permission.

disperse further than their natal lake or pond. Statoblasts, on the other hand, are resistant to dessication and therefore can survive overland dispersal. They have small hooks that facilitate their attachment to feathers or fur (Figure 4.4) and there is evidence that they can be transported 600 km or further on migratory waterfowl (Freeland et al., 2000). Other examples of species whose dispersal mechanisms vary throughout their life cycles are shown in Table 4.5.

Another important reason for relatively complex dispersal mechanisms in plants, fungi and invertebrates is the ability that some of them have to disperse through time. Many plants have seed banks, which are formed when seeds become covered up by substrate and enter a period of dormancy that, depending on the species, may last anywhere from a couple of years to 100 years or longer. Delayed germination causes genotypes that may have been absent for some years to be reintro-duced into a population, resulting in temporal gene flow. In one extreme example,

Table 4.5 Examples of species in which dispersal mechanisms vary throughout their life cycles


Pre-adult dispersal

Adult dispersal

Stonefly (Peltoperla

Local: larvae that crawl or

Local and long-distance: flying


are swept by current

or wind-borne


Local: larvae ingested by

Local and long-distance:

(Microphallus sp.)

freshwater snail host

dispersed by waterfowl hosts

(Potamopyrgus spp.)

Water flea (Daphnia

Local and long-distance:

Local: swimming


eggs swept along by

currents. Long-distance:

eggs transported by

animal vectors, boats

Honey mushroom

Local and long-distance:

Local: underground hyphae

(Armillaria bulbosa)

spores blown by wind

Red seaweed

Local and long-distance:

Local and long-distance: floating

( Gracilaria gracilis)

male gametes and spores

torn thallus fragments

swept by current


Local and long-distance:

Local and long-distance: rafting

(Adalaria proxima)

larvae swept along by

on drifting algae

ocean currents

seeds from sacred lotus plants (Nelumbo nucifera) were recovered from a lake bed in the Liaoning Province, China. Although these were dated at between 1200 and 330 years old, the majority germinated and grew (Shen-Miller et al., 1995).

Some fungi produce propagules by asexual means that can remain dormant for several years. Sclerotia are multicellular propagules produced by the ascomycete fungi in the genus Sclerotinia, and these can remain viable in the soil for at least 4-5 years (Adams and Ayers, 1979). A number of invertebrate species, including some rotifers, bryozoans and copepods, have banks of diapausing eggs or stato-blasts that are analogous to seed banks and also result in temporal gene flow. In some of these species the eggs can remain viable for more than 300 years (Hairston, Van Brunt and Kearns, 1995). In other habitats, juvenile sugar-beet cyst nematodes (Heterodera schachtii) can survive encysted in the soil for many years, and the nematode Caenorhabditis elegans responds to unfavourable environmental conditions by diapausing as a dauer larva that can last for several times the normal life span. Many insect species are also capable of prolonged diapause, such as the Yucca moth Prodoxus y-inversus, whose larvae developed into adults following a 30-year diapause in laboratory winter-like conditions (Powell, 2001).

The complexity of dispersal mechanisms, and the small size of many dispersing units (e.g. pollen, spores) in plants, fungi and invertebrates have made direct observation of movement very difficult in these taxonomic groups. Even when estimates of dispersal have been quantified using direct methods such as seed traps, they may bear little relation to gene flow. The contribution of molecular data to our understanding of dispersal and gene flow in these species has therefore been particularly noteworthy, as we shall see in the following sections that will focus on some of the most important factors that influence gene flow: dispersal ability, barriers to dispersal, reproductive mode, habitat patchiness, and interspecific interactions. In this chapter the emphasis will be on invertebrates, plants and microbes, but in Chapter 7 we shall add to our discussion by looking at patterns of dispersal in a number of vertebrate species.

Dispersal ability

Although dispersal does not always lead to gene flow, it is a necessary precursor and there is, as we would expect, a tendency for highly mobile species to show accordingly high levels of dispersal and gene flow. For example, one review estimated the mean maximum dispersal distance as 148.1 km in 77 bird species, compared with 74.6 km in 40 mammal species (Sutherland et al., 2000), a result that is not particularly surprising given the flight capacity of birds. Similarly, in this chapter's earlier section on indirect estimates of gene flow, we saw that a correlation between dispersal ability and Nem has been used to defend the accuracy of gene flow estimates that have been obtained in this way. Although this may be considered a somewhat circular argument, it gains support from an additional correlation between dispersal ability and gene flow that is based on more complex patterns of dispersal: isolation by distance.

Isolation by distance

So far we have been comparing dispersal ability with gene flow estimates that were based on a single (usually an average) FST value. Reducing complex patterns of gene flow to a single variable is logistically attractive but it does mean that a lot of detailed information about dispersal patterns will be lost. For example, in many species the amount of gene flow between populations is inversely proportional to the geographical distances between them because individuals are most likely to disperse to nearby sites. This is known as isolation by distance (IBD; Wright, 1943), and is most commonly assessed by regressing log-transformed estimates of gene flow between pairs of populations against the appropriate log-transformed geographic distances. The significance of this relationship can be assessed using a Mantel test, which tests for a correlation between genetic and geographical distances. The Mantel test is appropriate for this comparison because it does not assume that the population pairwise comparisons are independent. The slope and the intercept of the IBD regression can be used to test the strength of the relationship.

Isolation by distance No isolation by distance

Isolation by distance No isolation by distance

Was this article helpful?

0 0

Post a comment