This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Gene flow in marine invertebrate may play a critical role on evolution of populations because of characteristics of marine environment: open and connected by water. This is because different gene flow will give rise to different level of genetic structure among populations, and the level of genetic structure and population connectivity will eventually decide evolutionary path of populations. Two distinct developmental modes, planktotrophy and lecithotrophy are observed during life cycle of marine invertebrates, and they are assumed to affect differently on gene flow, with higher gene flow being expected for planktotropy. However, current evidences show that the difference of larval developmental modes alone is not responsible, but that variation in gene flow in marine invertebrates occurs by the cooperation of the biology of marine invertebrates and ocean environments. Therefore, more studies will be required to understand the biology of marine invertebrates, especially larval behavior and oceanographic characters.
Life cycle of marine species is diverse and complicating processes are involved from spawning to settlement. However, most marine invertebrates have larval stage in their life cycle, and most of the variation in gene flow among populations appears to be caused during this stage. This is because the common larval stages, which are planktotrophy and lecithotrophy show different larval nature, in which planktotrophic has longer life span and more dispersal capability than lecithotrophy. It also appears that the combination of other factors, including but not limited to oceanographic and environmental factors and larval behavior may be responsible for gene flow in more fine-scale. Gene flow among populations will then design population connectivity, which will play a role on evolution of populations, therefore in order to estimate the history and future of populations, accurate measuring population connectivity is required using various direct and indirect methods. Here, therefore life cycle and larval nature of marine invertebrate will be explained and their role on larval dispersal coupled with other oceanographic and environmental factors will be discussed, and then, several methods measuring gene flow and population connectivity will be touched on. Lastly, current studies about marine gene flow and connectivity and future studies will be discussed.
Life history diversity of marine invertebrates
Many marine invertebrates maintain their populations through the processes of gamete production, larval development, larval dispersal, settlement, metamorphosis and growth.
Production of larvae
For spawning and fertilization during the reproductive cycle in marine invertebrates, three methods are involved: copulation, pseudo-copulation and free spawning. Marine species using copulation with copulatory structure produce fewer eggs and often protect them during their development. Pseudo-copulation in which gametes are released into slime is used for other marine species to increase their protection of eggs. Still a large number of marine invertebrates fertilize through free spawning, taking advantage of sea water.
After fertilization, most embryos develop into larval stage that involves two modes of larval development, lecithotrophic and planktotrophic development. Lecithtropic larvae depend on the yolk in a relatively large egg for nourishment, and spend a few hours to a day or so moving limited distance. On the contrary, planktotrophic larvae feed on plankton, drifting for one to several weeks, and usually have a long-distance dispersal capability. The geographical composition of the two larval stages shows gradual changes from the Arctic or Antarctic to the tropics, since the percentage of planktotrophic development is increasing toward the tropics (Fig. 1.), and this may be attributed to the availability of phytoplankton and the longer periods of phytoplankton production at higher water temperature in tropical area than the Arctic and the Antarctic area (Thorson 1950). Thanks to their difference, the two larval stages are expected to play a different role on evolution of populations. Curiously, marine invertebrates that produce both modes of larval development, lecithotrophy and planktotrophy have been reported, and the presence of more than one mode of development in the same sexually reproducing species is termed poecilogony (Giard 1905).
Cases of poecilogony have been reported, especially in mollusks and polychaetes. Since the difference between lecithotrophy and plaktotrophy, specifically in their behavior, life span and dispersal capability, causes various ecological and evolutionary consequences, poecilogony may provide insight into the ecological importance of selecting alternative strategies (Levin and Bridges 1994, Ellingson and Krug 2006), and the developmental mechanisms (Marsh and Fielman 2005). However, poecilogony is still in its infancy for study and identifying real cases seems complicating. After their comprehensive reviews on papers reporting this phenomenon, Hoagland and Robertson (1988) reported that they found no real case of poecilogony and argued that most cases was not supported by sufficient data, or it turned out that they cryptic species. More thorough and careful studies will be required for confirmation of two modes using various methods such as breeding experiments and genetic data. However, poecilogony may be more common than might be thought, because molecular or inter-breeding data is available in polychaetes (Gibson et al. 1999, Morgan t al. 1999, Schulze et al. 2000, Gibson and Gibson 2004) and in opisthobranch molluscs (Gibson and Chia 1989; Krug 1998; Krug 2005).
Studies suggest that transitions between the developmental stages have occurred frequently in the history of many taxa (Hadfield et al. 1995, Hart 2000, Jeffrey et al. 2003, Meyer 2003, Collin 2004). Although planktotrophy appears to be earlier mode than non-planktotrophy (Hansel 1982, Wray 1995), the evolutionary forces of the transition towards nonfeeding larvae are unclear. However, several evolutionary factors were suggested in relation to the evolution of nonfeeding development (Chester 1996, Ellingson and Krug 2006). Ellingson and Krug (2006) thought that lecithotrophy could be used to reduce larval mortality caused by offshore transportation or food shortage, and lecithotrophy could be an alternative when planktotrophic larvae are blocked from returning to their habitat. Besides, the nutritional regime of adults can affect the developmental mode in that adults in healthy condition would produce larger eggs that will likely develop into lecithotrophic larvae, while starved adults would release smaller eggs and they will develop into planktotrophic larvae because of their food shortage (Chester 1996). More studies on poecilogony will provide insight into what are the factors selecting alternative development mode, shaping life-history evolution of marine species.
Larval dispersal and settlement
During the pelagic larval stage, larvae feed on phytoplankton, search for a suitable settlement site and grow, giving the larvae opportunities to disperse. Some larvae are able to delay their final metamorphosis for days to weeks until they find a settling place. If metamorphose occurs at an unsuitable settlement place, larvae will not survive. Because they are required to find a suitable settlement site within a limited time, many invertebrate larvae appear to have evolved to ensure their successful settlement. Once larvae manage to reach to a settling site, more problems will be waiting for their successful settlement and recruitment into the population. For example, filter feeders can eat larvae at settlement sites. In addition, settlers may be swept out to deep sea by waves, or may be dried out because of too low tidal height. There are competition and predation as well.
Although other factors including competition, and predation may be important influences for the structures and dynamics of populations, the consequences of variation in numbers and timing of offspring arriving into the habitat will be crucial, and variations in recruitment occurs in marine invertebrates, because of many processes involved in their life history, such as production and dispersal of larvae, mortality while dispersing, and the settlement, growth and survival of larvae (Underwood and Keough 2001), and "supply-side ecology" relate variation in larval input and the size of adult populations (Caley et al. 1996, Hughes et al. 2000). For example, Fig.2. may show well how variation in recruitment is affected by supply-side, as they show that species with long planktotrophic larval stage tend to produce a really large number of eggs and thus more fluctuating population size from year to year than species with non-pelagic development, and the terrestrial (Thorson 1950).
The risk of fertilization failure always exists in marine invertebrates. While direct transfer of sperms from males to females can ensure high fertilization success, that of free spawning is affected by several factors. Sperms and eggs may have a limited chance of mixing together because of low population densities and the distance between males and females, and water movement may dilute gamete concentration to reduce the probability of fertilization. Different timing of egg and sperm release minimizes the success, as well. In addition, gametes spawned from many species at the same time at the limited place can be problematic for species specific fertilization. Sperm age also can affect the success rate in that sperm of most invertebrates have relatively short life span, and thus aging sperm may have limited fertilizing ability. However, some marine species with appropriate behavior seems to be maximizing the fertilization success. Firstly, most marine species will produce from several thousands to millions of eggs per female in their lifetime, whereas most terrestrial invertebrates produce a few hundred to a few thousand eggs (Fig. 2). Several marine invertebrates mimic copulatory organs by spawning eggs and sperm in a mass of protected slime (pseudo-copulation), and some will not lay their gametes unless they are quite close to one other. Even so, large number of marine invertebrates will still spawn their eggs and sperm freely in the water, so waste of gametes and low fertilization success rate will be inevitable.
The larval waste is an important part from the supply-side point of view, since the size of successful recruitment will depend on the degree of larval waste. Because normally lecithotropic larvae are short-lived and depend on internal food sources, larval waste is therefore found mainly in planktotrophy, and the number of larvae recruitment at a population will be affected by several important processes, which include the period of larval dispersal and the mortality during their dispersal. Since the majority of the planktotrophic larvae feed on phytoplankton organisms, food limitation or bad quality of food may kill larvae. Temperature may also play a role on larval waste in that low temperature may lengthen the pelagic life or postpone metamorphosis in some species (Nelson 1928, Korringa 1940), by which the larvae are exposed to predation. Ocean currents can cause a large larval waste as well by carrying larvae from their coastal habitat to deep sea. Above all, predation by other organisms will be a critical factor for the larval waste, because even all other factors mentioned above can eventually extend larval stage and can expose larvae to more chance of predation.
More studies will be needed to understand the degree of larval waste marine invertebrates experience, and whether or not there may be any other alternatives than massive larval waste, but Reddingius and den Boer (1970) suggested that all populations that retain only locally may have a high risk that their habitat will become unsuitable and the populations may become extinct, while long larval dispersal will increase the chance of finding a suitable habitat, when there are always chances of environmental changes, such as high temperature, waves and winds, for habitats to become unsuitable for their residents. Examplesâ€¦.
However, evidences show that some larval may have evolved to avoid massive larval waste. For example, it appears that many species take advantages of currents to return to local areas of their parents (Paris and Cowen 2004). Besides, some larvae may not be critically affected by food shortages (Olson and Olson 1989). Moreover, asexual reproduction is used for larvae to increase the chance of their survival (Knott et al 2003, Vickery and McClintoch 2000, Vaughn Sgtrathman 2008). Poecilogony might be an important method as well (Chester 1996).
Most larvae need to be transported by currents into appropriate habitats for settlement. During the movement of larvae by currents, they may be swept to inappropriate place, causing variation in larval supply (Gaines and Bertness 1992). Therefore, oceanographic features will play a critical role on larval dispersal and waste. Flows in nearshore differ from those of deep sea, because of the environmental features, such as shallow depths, shoreline barriers, and fresh water inputs, and thus flows in nearshore appear to be more complex than those of deep-ocean. Therefore, coastline topographic features may affect flows in the nearshore, causing complex flows in a smaller scale (Okubo, 1994). Although most major currents such as the Gulf Stream, the California Current, and East Australian Current tend to be steady, currents can be also affected by many factors, such as environmental changes (Connolly and Roughgarden 1999), the interaction between currents (e.g., Yasuda et al. 2009), and shapes of coastal line (e.g., Sotka et al., 2004). Winds and tidal bores play a role on larval transport as well (e.g., Dudas et al. 2009), and temperature may be a factor for larval transportation (O'Connor et al. 2007). Other species such as algal beds can also change larval dispersal around shoreline (Birrell et al. 2008). Moreover, disturbance of water column stratification-different layers of water column may affect diversity of biological communities (Striebel et al. 2010). Striebel et al. (2010) suggest that the availability of the resources for phytoplankton, such as light and nutrients depends on the depth of water column, and marine phytoplankton community is affected by disturbances of water stratification. Therefore, the phytoplankton community diversity will affect dispersal of larvae that rely on phytoplankton for food.
It has been generally assumed that larvae are rather passively transported than actively swimming (Roberts, 1997). However, increasing number of studies show that some marine species of larvae are able to swim faster than previously expected, and swimming ability of larvae is an important part for larval movement (e.g., Phillips and Sastry, 1980, Leis and Stobutzki, 1999; Pizarro and Thomason 2008). An example is vertical swimming ability. Most coastal water is stratificated and each layer may have different characteristics, such as flow directions and speeds, therefore larvae may vertically move into different layer of the water column and being transported to different directions or at different speeds (e.g. Schmalenbach and Buchholz 2010).
Larval also have sensory capabilities to respond to habitat cues (review of Sponaugle et al. 2002), by which larvae may swim actively between water strata to enhance their successful transport to the target area. Zimmer-Faust and Tamburri (1994) reported that naturally occurring waterborne compounds can stimulate the settlement oyster (Crassostrea virginica), and Hadfield and Koehl (2007) found that larvae of the nudibranch Phestilla sibogae rapidly responded to dissolved chemical cues from Porites compressa, suggesting that larvae of this species may be able to increase the chance of dispersing to the substratum by chemical settlement cues.
Gene flow and Population connectivity
Different life history and reproductive strategies in marine species can cause variation in dispersal capability, and dispersal differences may produce different level of genetic exchange between populations within a species, since dispersal generally play a homogenizing effect while restricted gene flaw causes differentiation among populations (Palumbi 1994). Marine populations may be open, in which local recruitment is decided by settlement after pelagic larval dispersal, while separate patches will be closed if gene flow is limited between populations (Hixon et al. 2002). For example, low genetic structure (probably more open structure) was found within sea star (Pisaster ochraceus) populations from Western USA cost and B.C., Canada (Harley et al. 2006), and an example of relatively closed population was observed in a study in many Penaeid shrimp speices along Brazilian coastline, where high genetic structures among the species were found (Gusmão et al. 2005). Therefore, the role of larval dispersal in population dynamics will depend on whether the new recruits are from near or far distance, since larvae from different populations may be genetically different, having variation in their settlement success (e.g. Hamilton et al. 2008) and, populations located further distance will likely have more different genetic composition than populations closely located. Limited gene flow or low level of genetic connectivity between populations may cause isolation that may play an important role evolutionary, regarding speciation and extinction (Palumbi 1994). Using COI and phylogenial methods, Palumbi et al. (1997) found high level of genetic differentiations (Fst=0.145-0.389) within four Pacific species of Sea Urchins (Echinometra), and suggested that the emergence of these species and their population structure may be caused by isolation and speciation.
Measuring larval dispersal distance
Measuring larval dispersal distance, especially in marine species is a very difficult task, because most marine species spawn millions of propagules that are too small to be seen with the naked eyes. Besides, in the case of planktotrophic development especially, they spend a long period time in the plankton for observers to follow them. Therefore, direct methods en used such as tagging propagules (e.g., Almany et al. 2007) and direct observation of the individual propagules are inefficient and give very limited information about larvae dispersal. For this reason, for many marine species, dispersal distance scales have been linked to the time larvae spend in the plankton, the pelagic larval duration (PLD) (Shanks et al. 2003), and PLD has been used in estimating dispersal. Yet, simply relating PLD to larval dispersal distance may be insufficient, because larvae do not seem to disperse in a passive way, but their dispersal is affected by several other factors, such as timing of spawning, type of larvae, larval behaviors, propagule duration, and local hydrodynamics (Shanks 2009). Geochemical (Thorrold et al. 2002, Thorrold et al. 2007), genetic markers (Pascoal et al. 2009), and biophysical models (Siegel et al. 2008) will be the alternative methods of estimating larval dispersal and population connectivity. Variation in certain environmental conditions such as temperature, salinity and water chemistry may generate natural geochemical tags in shells, otoliths and statoliths, and these tags can be used for tracking larvae dispersal (Becker et al. 2007), and fluorescent tags and radioactive isotopes are used for artificial geochemical tags (Thorrold et al. 2002). However, limitation exists to apply these geochemical markers in that normally little variation in natural geochemical tags is observed between populations (i.e., Gillanders et al. 2001), and thus characterizing natural geochemical tags of all populations interested is required. For artificial tags, tagging a large proportion of larval in an area before dispersal and intensive sampling at their potential arriving areas is required. Biophysical models can be another approach as well, since with the advance of computer and software, larval dispersal can be estimated by biophysical models. This can be done by calculating dispersal kernels-larval settlement probability distribution by physical (e.g., water column structure, water flows, and water velocity) and biological parameters (e.g., timing of larval release, vertical movement, and PLD), but with many drawbacks thanks to the lack of understanding about larval nature such as larval mortality and settlement (Metaxas and Saunders 2009). Genetic markers can be an alternative.
The development of polymerase chain reaction (PCR), various kinds of genetic markers, and sophisticated computer programs all have increased the importance of population genetics field in biology. DNA is easier to extract, treat and store, and more informative than RNA. Using DNA requires little sample than using RNA, thanks to PCR. Different kinds of markers can be used for different purposes (e.g., Genealogy Vs. frequency, single-locus vs. multi-locus). Therefore, decision on which genetic markers to use is most important for a population genetic survey (see table 1).
Allozyme markers, which are codominant and inexpensive have been widely applied to investigate population structure in marine invertebrates (e.g., Gomez-Uchida et al. 2010) especially when DNA based markers were not very developed. However, these markers provide often limited variation and thus limited genetic data, requiring high-quality samples. Besides, it is assumed that allozymes likely experience some sort of selection, because they are the products of functional genes (Riginos et al. 2002). Therefore, DNA based markers are becoming more popular.
RAPD (Random amplified polymorphic DNA), AFLP (Amplified fragment length polymorphism) and microsatellite markers are PCR-based DNA markers, and RFLPs and sequencing are non PCR-based markers. RAPDs and AFLPs require many anonymous primer sets, and thus they are technically easy to apply. RFLPs are DNA fragments digested by restriction enzymes, followed by elctrophoresis. Sequencing that detects variations to nucleotide (A, G, C, and T) can be conveniently used for most organisms and be used for phylogeography and systematics. Microsatellites that are widely distributed throughout the genome have advantages over other markers (Powell et al. 1996), in that they are highly polymorphic compared to other markers (Rowe et al., 1997). Besides, microsatellites are codominant markers and can have many alleles, being more informative, while RAPDs and AFLP are dominant markers, and sequencing provide only two alleles. Although developing microsatellites is labor intensive and expensive, they can be applied to other close species once developed (e.g. Keever et al. 2008). RAPD and AFLP also have other technical issues of scoring errors and reproducibility requiring special care for their usage and interpretation. Nevertheless, multilocus markers (RAPD and AFLP) can still be useful for certain studies such as gene mapping and quantitative traits (e.g., Yu and Guo 2003).
Population genetics to population genomics
Population genetics requires understanding the genetic basis of population adaptation, since adaptation of the populations and their genetic structure are characterized by evolutionary factors, environment and population gene pools (Krutovsky 1988). Many genetic markers (microsatellites, RAPD, and AFLP) can be used for adaptive gene identification (Guinand et al. 2004). This is because of the fact that some of the markers linked to genes undergoing natural selection behave differently from other neutral markers (called 'outliers'). These outliers can be detected by the FST outlier test (Beaumont and Nichols, 1996), because of their exceptionally high or low genetic differentiation (as measured by FST) than that of other neutral genes.
However, using traditional molecular genetic markers has limitations for understanding the process of adaptation, because they are likely to be neutral and are only indirect methods for these kinds of studies. In this respect, genomic approaches at the genome-wide level will be necessary to study the genes responsible for adaptive traits in populations to understand population adaptation and evolution (Luikart et al. 2003). For this approach, single-nucleotide polymorphisms (SNPs) or Expressed Sequence Tags (ESTs), which normally represent expressed functional genes have been widely used to estimate genetic structure (e.g., Tanguy et al. 2008).
Population Genetic Concepts
Genetic characteristic of populations is built by allele frequencies that result from evolutionary factors such as random genetic draft, selection, mutation and gene flow. Mutation and gene flow can homogenize populations, whereas genetic draft and selection have diversifying effect. Two parameters, m (the genetically effective migration rate) and Ne (the genetically effective population size) are used for measuring gene flow (Slatkin 1985b) and these parameters are incorporated into the most popular formula for gene flow estimation, Fst=1/1+4Nem (Wright 1931). Fst ranges from 0 to 1, and theoretically the lower Fst, the higher gene flow.
In practice, however, special care should be taken to interpret data in using this formula with genetic markers for several reasons. The formula is based on an island model of dispersal in which all the subpopulations have equal population size and equal migration probability. For those species living along coastlines, a stepping-stone model of dispersal will be more suitable. As mentioned genetic structure among populations can be affected by other evolutionary forces and they can bias the level of gene flow, unless all the forces are not at equilibrium within the populations (Felsenstein 1982). Besides, measuring low level of Fst may cause errors. When the two effects are not in equilibrium, estimated gene flow may be highly biased (e.g., Bohonak and Roderick 2001). Subpopulations reach their equilibrium with different time period and the time can be estimated by 1/2m+1/2Ne (Crow and Aoki 1984), so depending on migration rate and effective population size, current subpopulations may or may not be in their equilibrium. For example, high level of Fst between subpopulations can show either limited gene flow, or a recent founding event (McCauley 1993; Hutchison and Templeton 1999), and low level of genetic structure is estimated between subpopulations either because of high level of current gene flow or because of very recent isolation (Felsenstein 1982). Different markers may show different Fst values, as well.
DNA sequencing has empowered the ability to characterize genetic structure, because direct sequencing can detect substitutions at DNA level and has made it possible to infer the relationship between alleles. Interpreting the genealogical data requires coalescence theory which traces all alleles of a gene in a population to the most recent common ancestor (MRCA), and genealogies can be drawn into gene trees (Kingmen 1982). For example, when there is recent secondary contact of populations after a long period time of isolation, Fst will reveal high level of genetic structure. In this situation, Fst approaches do not show if the high level of genetic structure is because of secondary contact after isolation or sudden reduction of gene flow, while gene genealogies will draw reciprocal monophyly which shows a long period time of isolation (Cunningham and Collins 1994).
Self-recruitment is more common
Estimating the degree to which marine populations are maintained by self-recruitment or external sources is an interesting and difficult challenge for marine biologist today. The degree ranges from entirely closed (100% self-recruitment) to fully open populations depending on other sources for their recruitment. It may be generally assumed that long distance dispersal and connectivity is common among subpopulations because of pelagic larval stage, and that local marine populations are rather open to local and non-local sources for their recruitment. However, more evidences suggest that marine populations may be more closed than previously expected (see Swearer et al. 2002). Genetic methods of Perrin et al. (2004) identified significant population structure among sea star (Coscinasterias muricata) populations around New Zealand coast at the distance scale of both >1000km and 10-200km. Using chemical tags, Becker et al. (2007) found the evidence of self-recruitment within mussel species, Mytilus californianus and M. galloprovincialis in West cost of USA. Data of limited gene flow among populations is available in more fine-scale, which will the study of Banks et al. (2007), in which neighboring population were found to be genetically more different than more distant population between New Zealand Centrostephanus rodgersii populations. With 2D Advection-Diffusion-Mortality simulation model, Cowen et al. (2000) also found insufficient larval exchange, even between very closely located populations at typical current speeds (30 to 50 s-1).
Although mechanisms responsible for more limited gene flow and more self-recruitment than expected in diverse marine taxa are not well understood, some suggestions are available. Variation during larval stage (e.g., larval duration, mortality, metamorphosis) can be responsible for connectivity patterns (Becker et al. 2006). As mentioned before, larvae may be able to actively make vertical movement in water column, choosing different flow layer (Becker et al. 2006; Cowen et al. 2000), and larval ability to respond to settlement cues may play a role (Kingsford et al. 2002).Besides, intermingle of physical oceanography, coastal geography, and genetic structure will affect population connectivity in fine-scale (Banks et al. 2007).
Latitudinal gradient in gene flow
The earth is composed of 5 oceans, which are the Atlantic, Arctic, Indian, Pacific, and Southern Ocean, covering more than 70% of the Earth's surface, in which unlimited numbers of marine species are living through a broad geographic area. Therefore, it may be intereting to study the level of divergence among populations of different species and the related processes, especially when geographically different areas have different environmental characteristics, such as temperature and salinity. In this respect, estimating gene flows and comparing them among populations of many species broadly will be a starting point (e.g., Dawson 2001; Wares 2002).
Two larval developmental modes, pelagic and non-pelagic, are distributed along the ocean with there being poecilogony phenomenon in some taxa, but showing gradual increase of pelagic; gradual decrease of non-pelagic toward lower latitude (Fig. 1., Thorson 1950). Therefore, a question will be whether or not higher gene flow will be expected at lower latitude. Although it may be assumed their connectivity at lower latitude because of their long pelagic larval stage on currents, it should not be generalized because evidences suggest that development modes are not necessary related to larval dispersal (Young et al. 1997; Todd et al. 1998).
Next question is whether certain trends will be found along with latitudinal gradient for genetic structure among population within a species with the some developmental mode, since it has been suggested that larval development occurs faster in warmer waters in general (Strathmann 1974; Hoegh-Guldberg and pearse 1995). Only limited data is available for this kind of study. Using Nacella concinna with planktotrophic larval stage, Hoffman et al. (2010) found lower genetic structure among populations from Antarctic Peninsula than those from lower latitude, and more isolated populations were found at lower latitudes, when Kelly and Eernisse (2006) studied with 28 chiton species along the Pacific coast of North America.
The latitudinal pattern of poecilogony in terms of individual's selecting alternative developmental mode is also questionable, because if poecilogony is more common in some taxa, their choices will play an important role on larval dispersal. The study of Krug (2007) may provide a little insight into the phenomenon, in which they found larval ability to switch the developmental mode seasonally in Alderia, suggesting that environmental cues (temperature and salinity) could affect larval development
Gene flow among populations in many marine species seems limited, therefore more studies will be needed to find the causes. So,
Understanding larval nature including larval behavior (larval swimming ability and larval response to cues).
Understanding oceanographic features and environmental factors and estimating their link to larval dispersal.
Studying about larval developmental modes and poecilogony will provide insight into gene flow. So,
Estimating how common the phenomenon is in marine species and the pattern at global level.
Understanding the effect of environmental changes (increasing temperature and pollution) on poecilogony and thus on gene flow among populations.
Studying the effect of poecilogony on population evolution.
Measuring genetic structure more accurately will be required
Genetic data will be required in wider scale, with more samples, on a more regular basis.
Various genetic markers will be applied to complement other drawbacks.
More development will be required for other direct and indirect tools, such as tagging, modeling, and genomics to back up genetic data.