Cospeciation and rates of evolution in lice and birds: a molecular approach

PI: Rod Page (r.page@bio.gla.ac.uk)

Introduction

A key question in the study of coevolution is the tempo and mode of evolution of the interacting partners. For example, how old is the association between a given parasite and its host? Is the association an ancient one, reflecting a long and intimate interaction between the two organisms, or is it a recent event due perhaps to a parasite colonising a new host? Answering these questions requires the comparison of evolutionary trees (phylogenies) for host and parasite (Fig. 1).

Figure 1
Matching and mismatching pairs of host and parasite phylogenies. Matching (top) is evidence for host-parasite cospeciation; mismatching (bottom) is evidence for processes other than cospeciation.

To the extent that the trees match, host and parasite have cospeciated, that is, parasite and host speciate at the same time. Mismatches between the host and parasite phylogenies trees signals processes other than cospeciation, such as host switching, speciation by parasites independently of their hosts, and parasite extinction. Documenting the extent of cospeciation is both a fundamental step in interpreting the history of a host-parasite assemblage, and also provides a temporal framework for comparative analyses of the evolution of host and parasite.

Molecular phylogenies are an especially powerful means of testing hypotheses of cospeciation, particularly if the gene being analysed evolves at an approximately constant rate (i.e., shows a molecular "clock"). Given a molecular clock, it is possible to tease apart whether incongruence between host and parasite phylogenies is due to recent host switching, or the differential survival of old parasite lineages (Page 1994; Page, Clayton et al. 1996). Host-parasite systems are also particularly suited to studies of comparative molecular (Hafner and Page 1995). The amount of evolutionary change, d, between two species is the product of the rate of evolutionary change, r, and the time t since the two species had a common ancestor, hence d = rt. In order to compare rates of change in two or more different lineages we need to know the relative ages of those lineages. If t is unknown, then differences in the relative contribution of age and rate to relative evolutionary divergence cannot be established; difference in divergence might reflect difference in age between the two lineages, or different rates of evolution, or both.

Comparisons of rates of evolution among free-living organisms are hampered by the requirement for a good fossil record for all taxa being compared, in order to accurately determine the age of these taxa. Cospeciating hosts and parasites uniquely allow us to circumvent this difficulty because cospeciating taxa are, by definition, contemporaneous. This temporal link between the two clades can be exploited in studies of rates of evolution; in particular, it frees such studies from dependence on the fossil record, either partially by using the fossil record of, say, the host to calibrate the absolute rate of evolution of the parasite (Moran, Munson et al. 1993; Bandi, Sironi et al. 1995; McGeoch, Cook et al. 1995), or completely by determining the relative rate of evolution in host and parasite (Hafner, Sudman et al. 1994; Moran, van Dohlen et al. 1995; Page 1996).

Host and parasite are often taxonomically distant (e.g., birds and insects), with differing generation times, population sizes, and metabolic rates -- factors often invoked as explanations for differences in rates of molecular evolution (Martin and Palumbi 1993; Jermiin and Crozier 1994; Rand 1994; Dowton and Austin 1995; Mindell, Knight et al. 1996). Given sequences for homologous genes in both host and parasite, these hypotheses can be tested. Moreover, another advantage of host-parasite systems is the possibility of replicated hypothesis testing. Some host-parasite assemblages comprise suites of related parasite lineages associated with the same set of hosts over evolutionary time. This allows comparisons among lineages, all employing the same temporal control provided by cospeciation (Fig. 2).

Figure 2
Two parasite clades cospeciating with the same host. The two parasite clades provide replicate lineages of the same age as the corresponding host lineages.

These attributes of taxonomic disparity between host and parasite, and the temporal control provided by cospeciation make host-parasite assemblages ideal for investigating variation in rates of evolution among clades (Hafner and Page 1995; Page and Hafner 1996).

Cospeciation

Five requirements for a comprehensive phylogenetic study of cospeciation (Page, Clayton et al. 1996) are: 1) adequate alpha-taxonomy of the hosts; 2) robust host and parasite phylogenies; 3) exhaustive sampling of parasite clades; 4) molecular phylogenies from comparable genes, and 5) quantitative comparison of host and parasite phylogenies. The first two requirements are obvious prerequisites. Exhaustive sampling of parasites increases the chances that all extant parasite lineages will have at least one representative included in the sample. In poorly sampled parasite clades incongruence between host and parasite phylogenies due to the presence of a number of lineages that have suffered some extinction may be misinterpreted as being due to host switching (Page, Clayton et al. 1996).

Using homologous genes ensures that we have comparable units of evolutionary change. In many cases the taxonomic distance between the host and its parasite means that morphologically they share few homologous features. Without such shared features it becomes very hard to compare amounts of evolutionary change in hosts and parasites. Molecular data remove this difficulty by enabling us to compare homologous nucleotide sites across distantly related taxa and to express evolutionary using the same units (e.g., numbers of nucleotide substitutions per site).

Hafner et al. (1994) (Hafner, Sudman et al. 1994) reported an 11-fold difference in rate of substitution between pocket gopher lice and their mammalian hosts, which they attributed to the order of magnitude difference in generation time between the short-lived lice and their longer-lived hosts. However, reanalysis of these data (Page 1996) suggests that the disparity in rates, though real, is actually only 2-3 fold. Although there is some evidence for generation time effects on the rate of molecular evolution in birds (Mooers and Harvey 1994) within insects generation time is a poor predictor of relative rates of evolution; for example the long-lived honeybee shows much greater evolutionary divergence than the short lived Drosophila (Crozier and Crozier 1992).

Dowton and Austin (1995) (Dowton and Austin 1995) found an increase in the rate of mtDNA sequence evolution in parasitic wasps coincident with the adoption of a parasitic lifestyle. Such a lifestyle could increase the rate of sequence divergence, either though an increased frequency of founder events (Ohta 1987) or through increased selection pressure due to a "genetic arms-race" between host and parasite. Hawaiian Drosophila that have undergone founder events show more evolutionary divergence than their sister clade that have no history of founder events (DeSalle and Templeton 1988). Pocket gopher louse populations resident on different individual hosts show high levels of inter-host differentiation (Nadler, Hafner et al. 1990). This suggests that the lice may undergo founder events with each initial infection of a juvenile gopher. Swift lice (Dennyus hirundinis) are transmitted vertically and have small population sizes (mean of 2 lice per bird) (Lee and Clayton 1995). This combination of vertical transmission and small populations may account for the accelerated rate of sequence evolution observed in lice relative to both their hosts and other insects (Hafner, Sudman et al. 1994). While gopher and swift lice are on hosts that have limited physical contact, more social hosts (and/or more mobile lice) may promote horizontal transmission of lice (Ròzsa, 1996) increasing the effective population size which would, in turn, be expected to reduce the rate of evolution.

Bird Lice

Figure 3
Phylogeny of louse suborders with the host distributions indicated. The traditional grouping of chewing and biting lice (the "Mallophaga") is paraphyletic.

Lice (Insecta: Phthiraptera) have played a prominent role in the development of ideas on cospeciation and coevolution (Barker 1994; Paterson, Gray et al. 1995) (Page et al., in prep.). Lice are also particularly attractive parasites for study because several louse clades show significant levels of cospeciation with their vertebrate hosts (Paterson, Gray et al. 1993; Hafner, Sudman et al. 1994) . Lice are also relatively easy to collect, and can be obtained from birds without harming or killing the (Fowler and Cohen 1983). Four louse suborders occur on mammals and birds. Birds are host to the suborders Ischnocera and Amblycera (Fig. 3) and a single bird species may harbour both suborders and several genera.

Our preliminary work (Page et al., in prep.) has established that the rate of mitochondrial DNA cytochrome b divergence in Dennyus lice is substantially (4-5) times greater than in Drosophila (Fig. 4) emphasising the lack of a global mtDNA clock within insects (Crozier, Crozier et al. 1989). Hence calibrations for other insects for which there is fossil or biogeographic evidence available (e.g. Drosophila (Powell and DeSalle 1995)) are inapplicable to lice. Therefore, studies of the rate of louse evolution are best undertaken with reference to the hosts.

Figure 4
Comparison of cyt b protein sequence divergence between the swift louse (Dennyus hirundinis) and its common ancestor with Drosophila (from Page et al. in prep).

Comparisons of available louse DNA sequences with homologous sequences from their hosts show that lice evolve 2-3 times more rapidly than their hosts (Page 1996) (Page et al., in prep.) allowing differences between insect and bird molecular evolution to be explored. Basic population data relevant to molecular rates of evolution, such generation time and the numbers of lice infesting individual birds (i.e., louse population size) are available for a number of species (Fowler and Miller 1984; Fowler and Price 1987; Rózsa, Rékálsi et al. 1996).

Another approach to comparing rates of evolution that does not rely on the fossil record is the relative rate test, where the divergence of two ingroup species to a common outgroup taxon is measured. The ratio of these values gives the relative rate of evolution in the two species. This method is useful among relatively closely related taxa, but for more distant taxa (such as insects and birds) the method becomes sensitive to the ability to accurately estimate sequence divergence. For rapidly evolving genes such as mitochondrial sequences this becomes difficult at such levels of divergence.

Hypotheses

The following hypotheses will be tested:

Lice cospeciate with their hosts

Louse phylogenies constructed from mt- and nuclear DNA will be compared with host phylogenies and trees tested for statistically significant congruence using the program TREEMAP. This program will also be used to document the relative number of cospeciation, extinction and host switching (Page 1994) in the swiftlet-louse and seabird-louse associations. Our preliminary work shows that at least some louse speciation is associated with swiftlet speciation (Clayton, Price et al. 1996) (Page et al. in prep.).

Bird louse DNA evolves more rapidly than host DNA

Our preliminary data indicate this is the case for swiftlets and their lice. This will be tested for seabird lice, and any disparity in rates will be tested for generality over mitochondrial (cyt b and 12S rRNA) and nuclear (18S rRNA and CHD1 gene) DNA. The relative rate of evolution will be determined at different codon positions in cyt b, and with reference to secondary structure domains in ribosomal genes.

Lice evolve more rapidly than their nearest relatives (Psocoptera)

If the greater rate of molecular evolution in lice is concomitant with their parasitic life style then we would expect lice to have accumulated more evolutionary change than their sister taxon (the Psocoptera) since the time these taxa diverged from their common ancestor. This hypothesis will be tested using a relative rate test to compare divergence in louse 18S rRNA obtained in this study with the published psocopteran sequence (von Dohlen and Moran 1995).

Higher rate of DNA evolution in lice is not due to change in selection regime

A measure of the nature and intensity of selection on a protein coding gene is the ratio of the number of synonymous (ks) to non-synonymous (ka) substitutions. If the difference in rate between host and parasite is not due to selection then the ratio ka/ks in homologous genes is expected to be the same in both hosts and parasites. If the ka/ks ratio is significantly different then at least part of the disparity may be due to different selection regimes in lice and their free-living relatives.

The rate of molecular evolution among lice is constant

If the rate of louse molecular evolution is affected by host population structure and the mode of transmission of lice (i.e., vertical versus horizontal) then we would expect the rate of evolution to vary among louse lineages. In particular the more mobile Amblycera tend to show lower host specificity than the Ischnocera, consequently may be expected to have larger effective population sizes and a slower rate of molecular evolution than the Ischnocera.

Determining the rates of speciation and extinction in lice

Given a completely sampled clade and a molecular clock it is possible to estimate relative rates of speciation and extinction (Nee, Holmes et al. 1994). This project will result in a phylogeny for the bulk of the known swiftlet lice, and our preliminary cyt b data suggests that this gene evolves at an approximately constant rate within this clade. Estimates of speciation and exinction rates derived using Nee et al.'s method will be compared to estimates of the number of speciation and extinction events postulated by the reconstruction of the history of the swiftlet/Dennyus louse association.

Methods

Two groups of lice will be sampled: swiftlet lice (genus Dennyus) and seabird lice (principally Saemundsonia, Halipeurus and Austromenopon). These taxa have been chosen on the basis of their availability and/or ease of sampling, known taxonomy, and availability of host phylogeny.

Swiftlet lice

Previous taxonomic work on swiftlet lice (genus Dennyus) (Clayton, Price et al. 1996) has resulted in an extensive collection of frozen and alcohol preserved lice which will be available for this study. The region of louse cytochrome b homologous with the region sequenced in their hosts (Apodiiformes) (Lee, Clayton et al. 1996) has been successfully amplified and sequenced in RG's laboratory. Preliminary analyses (Page et al., in prep) suggest that the 500 bp obtained from a subsample of taxa is adequate to build a reasonable tree (most clades with bootstrap values > 80) for these lice, however one major clade is poorly resolved and will require more data. Furthermore, several important louse taxa in the collections have not been sequenced. Given the importance of exhaustively sampling the parasite clade for accurately reconstructing the history of the host-parasite assemblage (Page, Clayton et al. 1996) we propose to sequence the remaining louse taxa for which we have material. Our collaborator Dr Clayton plans a collecting trip to the Philippines in 1997 to obtain further lice of the distinctus-group and their hosts. This material will supplement the collections we already have available.

Seabird lice

Seabird lice are highly host specific; even when different seabirds share nesting burrows very rarely do lice stray from their preferred host (Furness and Palma 1992). Seabird (principally the Charadriiformes and the Procellariiformes) phylogeny is becoming increasingly well (Bjørklund 1994; Moum, Johansen et al. 1994; Austin 1996) and comparisons of a molecular procellariiform phylogeny and a morphology-based louse phylogeny found significant congruence between host and parasite (Paterson, Gray et al. 1993; Paterson, Gray et al. 1995). The Glasgow University Applied Ornithology Unit has a large, active group of ornithologists who regularly travel to seabird nesting sites (e.g., the Shetlands and the Azores) and have in the past collected lice from these birds (Furness and Palma 1992). Some 14 seabird taxa can be relatively easily captured from these localities. These taxa harbour several ischnoceran and amblyceran louse (Fowler and Miller 1984). Lice will be collected from the birds (which are unharmed by the procedure (Fowler and Cohen 1983) and will be subsequently released) and stored in 100% ethanol (Post, Flook et al. 1993). Additional lice will be obtained by collaborators.

Data

Molecular data are central to the proposal for two reasons: (1) to provide comparable characters in lice and their hosts; and (2) to provide adequate phylogenetic characters for the lice, which tend to be morphologically very conservative. Subspecies of Dennyus which were only discovered using multivariate morphometrics (Clayton, Price et al. 1996) are more divergent in their mtDNA sequences than many bird genera (Page, et al. in prep). This makes the use of molecular data vital for reconstructing louse phylogeny at this taxonomic level.

We propose to test the hypotheses outlined above using a range of genes for which we know either insect (Simon, Frati et al. 1994) or louse-specific (our own work) primers exist. If necessary, additional louse primers will be developed. For most hosts sequences of the relevant mitochondrial genes are available, either through our own work or in GENBANK. The genes to be sequenced in lice are:

Mitochondrial cytochrome b

We have preliminary data from this gene for Dennyus lice and propose to complete sequencing this gene using louse specific primers. Seabird lice will also be sequenced (no mitochondrial protein data is available for seabird lice).

Mitochondrial 12S rRNA

This gene is more conservative than cyt b and will allow us to determine whether the elevated rate of cyt b is typical of mtDNA as a whole. The more leisurely rate of evolution in 12S rRNA is expected to provide better resolution of deeper louse phylogeny. Limited unpublished 12S data has been obtained for some lice by Paterson (Paterson 1994) using universal primers. We will use these data to help design louse specific primers.

Nuclear 18S rRNA

This gene will be used to compare nuclear and mitochondrial divergence in lice. The availability of homopteran insect sequences (von Dohlen and Moran 1995), including one from a psocopteran which is the sister taxon to lice, will allow a relative rate test for louse divergence compared to related insects.

Nuclear CHD1 gene

One copy of this gene is present in the mouse, Drosophila and yeast (Stokes, Tartoff et al. 1996) so is likely to occur in the lice. It is also present as two independent copies (RG pers obs) one W linked and one Z linked in the birds (Griffiths, Dann et al. 1996; Griffiths and Korn submitted). PCR primers have been designed for the birds (Griffiths, Dann et al. 1996) and from the nuclear sequence data available can be designed for the louse without many problems. Comparative data from the two classes will be used to confirm conclusions made from the nuclear 18S rRNA study.

Sequences collected in this study will be deposited in GENBANK. Alignments will be placed on the "Taxonomy at Glasgow" Web server .

Analysis

Phylogenetic analysis will use parsimony PAUP (Swofford 1993), maximum likelihood (Felsenstein 1993) and spectral analysis (Hendy and Penny 1993). Tests for molecular clocks in both birds and lice will be performed using the likelihood ratio test implemented in PHYLIP DNAMLK. Cospeciation analysis will use the program TreeMap (Page 1994) and new software being developed by RDMP and his PDRA under NERC GR3/1A095.

The hypothesis of host-parasite cospeciation will be statistically tested by comparing the amount of similarity between seabird and lice phylogenies with that expected between random phylogenies (Page 1994). If the host and parasite trees are more similar than due to chance alone then that is evidence for cospeciation. Comparisons of rate of evolution in lice and birds will follow the protocol in Page (1996) (Page 1996).

References

Austin, J. J. (1996). "Molecular affinities of Puffinus shearwaters: preliminary evidence from mitochondrial cytochrome b gene sequences." Mol Phylogenet Evol 6: 77-88.

Bandi, C., M. Sironi, et al. (1995). "The establishment of intracellular symbiosis in an ancestor of cockroaches and termites." Proceedings of the Royal Society of London 259(B): 293-9.

Barker, S. C. (1994). "Phylogeny and classification, origins, and evolution of host associations of lice." International Journal for Parasitology 24: 1285-91.

Bjørklund, M. (1994). "Phylogenetic relationships among Charadriiformes: Reanalysis of previous data." Auk 111: 825-32.

Clayton, D. H., R. D. Price, et al. (1996). "Revision of Dennyus (Collodennyus) lice (Phthiraptera: Menoponidae) from swiftlets, with descriptions of new taxa and a comparison of host-parasite relationships." Systematic Entomology 21: 179-204.

Crozier, R. H. and Y. C. Crozier (1992). "The cytochrome b and ATPase genes of honeybee mitochondrial DNA." Molecular Biology and Evolution 9: 474-82.

Crozier, R. H., Y. C. Crozier, et al. (1989). "The CO-I and CO-II region of honeybee mitochondrial DNA: Evidence for variation in insect mitochondrial evolutionary rates." Molecular Biology and Evolution 6: 399-411.

DeSalle, R. and A. R. Templeton (1988). "Founder effects and the rate of mitochondrial DNA evolution in Hawaiian Drosophila." Evolution 42: 1076-84.

Dowton, M. and A. D. Austin (1995). "Increased genetic diversity in mitochondrial genes is correlated with the evolution of parasitism in the Hymenoptera." Journal of Molecular Evolution 41: 958-65.

Felsenstein, J. (1993). PHYLIP, Phylogeny Inference Package, University of Washington, Seattle.

Fowler, J. A. and S. Cohen (1983). "A method for the quantitative collection of ectoparasites from birds." Ringing and Migration 4: 185-9.

Fowler, J. A. and C. J. Miller (1984). "Non-haematophagous ectoparasite populations of procellariiform birds in Shetland, Scotland." Seabird 7: 23-30.

Furness, R. W. and R. L. Palma (1992). "Phithiraptera of petrels and skuas from Gough Island, South Atlantic Ocean." Seabird 14: 33-42.

Griffiths, R., S. Dann, et al. (1996). "Sex identification in birds using two CHD genes." Proc R Soc Lond 263: 1249-54.

Griffiths, R. and R. Korn (submitted). "A CHD1 gene is Z chromosome linked in the chicken Gallus domesticus." Gene.

Hafner, M. S. and R. D. M. Page (1995). "Molecular phylogenies and host-parasite cospeciation: Gophers and lice as a model system." Philosophical Transactions of the Royal Society, London 349(B): 77-83.

Hafner, M. S., P. D. Sudman, et al. (1994). "Disparate rates of molecular evolution in cospeciating hosts and parasites." Science 265: 1087-90.

Hendy, M. D. and D. Penny (1993). "Spectral analysis of phylogenetic data." Journal of Classification 10: 5-24.

Jermiin, L. S. and R. H. Crozier (1994). "The cytochrome b region in the mitochondrial DNA of the ant Tetraponera rufoniger: Sequence divergence in Hymenoptera may be associated with nucleotide content." Journal of Molecular Evolution 38: 282-94.

Lee, P. L. M. and D. H. Clayton (1995). "Population biology of swift (Apus apus) ectoparasites in relation to host reproductive success." Ecological Entomology 20: 43-50.

Lee, P. L. M., D. H. Clayton, et al. (1996). "Does behaviour reflect phylogeny in swiftlets (Aves: Apodidae)? A test using cytochrome b mitochondrial DNA sequences." Proceedings of the National Academy of Science, USA 93: 7091-6.

Martin, A. P. and S. R. Palumbi (1993). "Body size, metabolic rate, generation time, and the molecular clock." Proc Natl Acad Sci, USA 90: 4087-91.

McGeoch, D. J., S. Cook, et al. (1995). "Molecular phylogeny and evolutionary timescale for the family of mammalian herpesviruses." Journal of Molecular Biology 247: 443-58.

Mindell, D. P., A. Knight, et al. (1996). "Slow rates of molecular evolution in birds and the metabolic rate and body temperature hypotheses." Mol Biol Evol 13: 422-6.

Mooers, A. Ø. and P. H. Harvey (1994). "Metabolic rate, generation time, and the rate of molecular evolution in birds." Mol Phylogenet Evol 3: 344-50.

Moran, N. A., M. A. Munson, et al. (1993). "A molecular clock in endosymbiotic bacteria is calibrated using the insect hosts." Proceedings of the Royal Society of London 253: 167-71.

Moran, N. A., C. D. van Dohlen, et al. (1995). "Faster evolutionary rates in endosymbiotic bacteria than in cospeciating insect hosts." Journal of Molecular Evolution 41: 727-31.

Moum, T., S. Johansen, et al. (1994). "Phylogeny and evolution of the auks (subfamily Alcinae) based on mitochondrial DNA sequences." Proc Natl Acad Sci, USA 91: 7912-6.

Nadler, S. A., M. S. Hafner, et al. (1990). "Genetic differentiation among chewing louse populations (Mallophaga: Trichodectidae) in a pocket gopher contact zone (Rodentia: Geomyidae)." Evolution 44: 942-51.

Nee, S., E. C. Holmes, et al. (1994). "Extinction rates can be estimated from molecular phylogenies." Philos Trans R Soc Lond 344: 77-82.

Ohta, T. (1987). "Very slightly deleterious mutations and the molecular clock." Journal of Molecular Evolution 26: 1-6.

Page, R. D. M. (1994). "Parallel phylogenies: reconstructing the history of host-parasite assemblages." Cladistics 10: 155-73.

Page, R. D. M. (1996). "Temporal congruence revisited: Comparison of mitochondrial DNA sequence divergence in cospeciating pocket gophers and their chewing lice." Systematic Biology 45: 151-67.

Page, R. D. M., D. H. Clayton, et al. (1996). "Lice and cospeciation: A response to Barker." International Journal for Parasitology 26: 213-8.

Page, R. D. M. and M. S. Hafner (1996). Molecular phylogenies and host-parasite cospeciation: Gophers and lice as a model system. New uses for new phylogenies. P. H. Harvey, A. J. Leigh Brown, J. Maynard Smith and S. Nee. Oxford, Oxford University Press: 255-70.

Paterson, A. (1994). Coevolution of seabirds and feather lice: A phylogenetic analysis of cospeciation using behavioural, molecular and phylogenetic characters. Department of Zoology. Dunedin, University of Otago.

Paterson, A. M., R. D. Gray, et al. (1993). "Parasites, petrels and penguins: Does louse presence reflect seabird phylogeny?" International Journal for Parasitology 23: 515-26.

Paterson, A. M., R. D. Gray, et al. (1995). "Of lice and men: The return of the 'Comparative parasitology' debate." Parasitology Today 11: 158-60.

Post, R. J., P. K. Flook, et al. (1993). "Methods for the preservation of insects for DNA studies." Biochemical Systematics and Ecology 21: 85-92.

Powell, J. R. and R. DeSalle (1995). "Drosophila molecular phylogenies and their uses." Evolutionary Biology 8: 87-138.

Rand, D. M. (1994). "Thermal habit, metabolic rate and the evolution of mitochondrial DNA." Trends in Ecology and Evolution 9: 125-31.

Rosili, R. (1996). Systematics of skuas (Aves: Stercorariidae) with particular references to evidence from their feather lice (Insecta: Phthiraptera). Division of Environmental and Evolutionary Biology, IBLS. Glasgow, University of Glasgow: 239.

Rózsa, L., J. Rékási, et al. (1996). "Relationship of host coloniality to the population ecology of avian lice (Insecta: Phthiraptera)." Journal of Animal Ecology 65: 242-8.

Simon, C., F. Frati, et al. (1994). "Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers." Ann Entomol Soc Am 87: 651-704.

Stokes, D. G., K. D. Tartoff, et al. (1996). "CHD1 is concentrated in interbands and puffed regions of Drosophila polytene chromosomes." Proceedings of the National Academy of Science, USA. 93: 7137-42.

Swofford, D. L. (1993). PAUP, Phylogenetic Analysis Using Parsimony. Washington D.C, Laboratory of Molecular Systematics, Smithsonian Institution.

von Dohlen, C. D. and N. A. Moran (1995). "Molecular phylogeny of the Homoptera: A paraphyletic taxon." Journal of Molecular Evolution 41: 211-23.