Limits...
Aquatic Insects in Eastern Australia: A Window on Ecology and Evolution of Dispersal in Streams.

Hughes JM, Huey JA, McLean AJ, Baggiano O - Insects (2011)

Bottom Line: Studies that focus on contemporary timescales ask questions about dispersal abilities and dispersal behavior of their study species.In this paper we present a synthesis of connectivity studies that have addressed both these timescales in Australian Trichoptera and Ephemeroptera.We conclude with a number of suggestions for further work.

View Article: PubMed Central - PubMed

Affiliation: Australian Rivers Institute and Griffith School of Environment, Griffith University, Nathan QLD 4111, Australia. jane.hughes@griffith.edu.au.

ABSTRACT
Studies of connectivity of natural populations are often conducted at different timescales. Studies that focus on contemporary timescales ask questions about dispersal abilities and dispersal behavior of their study species. In contrast, studies conducted at historical timescales are usually more focused on evolutionary or biogeographic questions. In this paper we present a synthesis of connectivity studies that have addressed both these timescales in Australian Trichoptera and Ephemeroptera. We conclude that: (1) For both groups, the major mechanism of dispersal is by adult flight, with larval drift playing a very minor role and with unusual patterns of genetic structure at fine scales explained by the "patchy recruitment hypothesis"; (2) There is some evidence presented to suggest that at slightly larger spatial scales (~100 km) caddisflies may be slightly more connected than mayflies; (3) Examinations of three species at historical timescales showed that, in southeast Queensland Australia, despite there being no significant glaciation during the Pleistocene, there are clear impacts of Pleistocene climate changes on their genetic structure; and (4) The use of mitochondrial DNA sequence data has uncovered a number of cryptic species complexes in both trichopterans and ephemeropterans. We conclude with a number of suggestions for further work.

No MeSH data available.


Related in: MedlinePlus

(A) Map of Conondale (north) and Lamington (south) regions; (B) Cytochrome Oxidase 1 haplotype networks for three species sampled in both north and south regions [29-31]; (C) We used a hierarchical Approximate Bayesian Computation method, analyzed with MTML-msbayes [38], to estimate the number of divergence events (Ψ) that can explain the shared pylogeographic break in these three co-distributed taxa. Within the msbayes pipeline, we simulated 1,000,000 datasets (parameterized by random draws from predefined ranges) and calculated 23 different summary statistics from each dataset to create a prior distribution. The same summary statistics were calculated for our observed datasets (based on the CO1 region of the mtDNA molecule) and compared to the simulated summary statistics to generate posterior distributions for the variance of τ (time since divergence), average τ and Ψ. The posterior distributions shown here for E(τ) and Ψ were estimated using the local regression method. For the histogram describing Ψ, the red bars show the posterior distribution while the black bars show the prior distribution. For a more detailed explanation of msbayes see [12,38,39].
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4553437&req=5

f3-insects-02-00447: (A) Map of Conondale (north) and Lamington (south) regions; (B) Cytochrome Oxidase 1 haplotype networks for three species sampled in both north and south regions [29-31]; (C) We used a hierarchical Approximate Bayesian Computation method, analyzed with MTML-msbayes [38], to estimate the number of divergence events (Ψ) that can explain the shared pylogeographic break in these three co-distributed taxa. Within the msbayes pipeline, we simulated 1,000,000 datasets (parameterized by random draws from predefined ranges) and calculated 23 different summary statistics from each dataset to create a prior distribution. The same summary statistics were calculated for our observed datasets (based on the CO1 region of the mtDNA molecule) and compared to the simulated summary statistics to generate posterior distributions for the variance of τ (time since divergence), average τ and Ψ. The posterior distributions shown here for E(τ) and Ψ were estimated using the local regression method. For the histogram describing Ψ, the red bars show the posterior distribution while the black bars show the prior distribution. For a more detailed explanation of msbayes see [12,38,39].

Mentions: The upland habitats (all above 200 m) where T. palpata and B. narilla occur are separated in southeast Queensland, by a lowland area of about 100 km, which may have presented a significant barrier to dispersal during glacial periods because it was dryer and even more inhospitable than it is currently. Both species occur in the Conondale, D'Aigular, Blackall Ranges north of Brisbane and in the Lamington and Great Dividing Range south and west of Brisbane (Figure 3). These upland areas have been labeled a “mesothermal island archipelago in a sea of subtropical lowlands” [37], and it was hypothesized that mitochondrial sequences would show evidence for historical isolation of these upland islands. Both studies used a fragment of the cytochrome oxidase I gene. As predicted there were highly significant differences in genetic composition between these two regions, suggesting that currently dispersal is very restricted (T. palpata FCt = 0.10, p < 0.01, Φct = 0.11, p < 0.01; B. narilla FCt = 0.11, p < 0.001, Φct = 0.17, p < 0.001). For both species there were two clades. However, each clade was not entirely restricted to the northern and southern regions. McLean et al. [30] hypothesized that for B. narilla this may be due to a recent range expansion. However, distinguishing between gene flow and incomplete lineage sorting with one gene is difficult.


Aquatic Insects in Eastern Australia: A Window on Ecology and Evolution of Dispersal in Streams.

Hughes JM, Huey JA, McLean AJ, Baggiano O - Insects (2011)

(A) Map of Conondale (north) and Lamington (south) regions; (B) Cytochrome Oxidase 1 haplotype networks for three species sampled in both north and south regions [29-31]; (C) We used a hierarchical Approximate Bayesian Computation method, analyzed with MTML-msbayes [38], to estimate the number of divergence events (Ψ) that can explain the shared pylogeographic break in these three co-distributed taxa. Within the msbayes pipeline, we simulated 1,000,000 datasets (parameterized by random draws from predefined ranges) and calculated 23 different summary statistics from each dataset to create a prior distribution. The same summary statistics were calculated for our observed datasets (based on the CO1 region of the mtDNA molecule) and compared to the simulated summary statistics to generate posterior distributions for the variance of τ (time since divergence), average τ and Ψ. The posterior distributions shown here for E(τ) and Ψ were estimated using the local regression method. For the histogram describing Ψ, the red bars show the posterior distribution while the black bars show the prior distribution. For a more detailed explanation of msbayes see [12,38,39].
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4553437&req=5

f3-insects-02-00447: (A) Map of Conondale (north) and Lamington (south) regions; (B) Cytochrome Oxidase 1 haplotype networks for three species sampled in both north and south regions [29-31]; (C) We used a hierarchical Approximate Bayesian Computation method, analyzed with MTML-msbayes [38], to estimate the number of divergence events (Ψ) that can explain the shared pylogeographic break in these three co-distributed taxa. Within the msbayes pipeline, we simulated 1,000,000 datasets (parameterized by random draws from predefined ranges) and calculated 23 different summary statistics from each dataset to create a prior distribution. The same summary statistics were calculated for our observed datasets (based on the CO1 region of the mtDNA molecule) and compared to the simulated summary statistics to generate posterior distributions for the variance of τ (time since divergence), average τ and Ψ. The posterior distributions shown here for E(τ) and Ψ were estimated using the local regression method. For the histogram describing Ψ, the red bars show the posterior distribution while the black bars show the prior distribution. For a more detailed explanation of msbayes see [12,38,39].
Mentions: The upland habitats (all above 200 m) where T. palpata and B. narilla occur are separated in southeast Queensland, by a lowland area of about 100 km, which may have presented a significant barrier to dispersal during glacial periods because it was dryer and even more inhospitable than it is currently. Both species occur in the Conondale, D'Aigular, Blackall Ranges north of Brisbane and in the Lamington and Great Dividing Range south and west of Brisbane (Figure 3). These upland areas have been labeled a “mesothermal island archipelago in a sea of subtropical lowlands” [37], and it was hypothesized that mitochondrial sequences would show evidence for historical isolation of these upland islands. Both studies used a fragment of the cytochrome oxidase I gene. As predicted there were highly significant differences in genetic composition between these two regions, suggesting that currently dispersal is very restricted (T. palpata FCt = 0.10, p < 0.01, Φct = 0.11, p < 0.01; B. narilla FCt = 0.11, p < 0.001, Φct = 0.17, p < 0.001). For both species there were two clades. However, each clade was not entirely restricted to the northern and southern regions. McLean et al. [30] hypothesized that for B. narilla this may be due to a recent range expansion. However, distinguishing between gene flow and incomplete lineage sorting with one gene is difficult.

Bottom Line: Studies that focus on contemporary timescales ask questions about dispersal abilities and dispersal behavior of their study species.In this paper we present a synthesis of connectivity studies that have addressed both these timescales in Australian Trichoptera and Ephemeroptera.We conclude with a number of suggestions for further work.

View Article: PubMed Central - PubMed

Affiliation: Australian Rivers Institute and Griffith School of Environment, Griffith University, Nathan QLD 4111, Australia. jane.hughes@griffith.edu.au.

ABSTRACT
Studies of connectivity of natural populations are often conducted at different timescales. Studies that focus on contemporary timescales ask questions about dispersal abilities and dispersal behavior of their study species. In contrast, studies conducted at historical timescales are usually more focused on evolutionary or biogeographic questions. In this paper we present a synthesis of connectivity studies that have addressed both these timescales in Australian Trichoptera and Ephemeroptera. We conclude that: (1) For both groups, the major mechanism of dispersal is by adult flight, with larval drift playing a very minor role and with unusual patterns of genetic structure at fine scales explained by the "patchy recruitment hypothesis"; (2) There is some evidence presented to suggest that at slightly larger spatial scales (~100 km) caddisflies may be slightly more connected than mayflies; (3) Examinations of three species at historical timescales showed that, in southeast Queensland Australia, despite there being no significant glaciation during the Pleistocene, there are clear impacts of Pleistocene climate changes on their genetic structure; and (4) The use of mitochondrial DNA sequence data has uncovered a number of cryptic species complexes in both trichopterans and ephemeropterans. We conclude with a number of suggestions for further work.

No MeSH data available.


Related in: MedlinePlus