Research Update TRENDS in Ecology & Evolution Vol.17 No.8 August 2002 347

Research News Out of Gondwana: the origin of Scott V. Edwards and Walter E. Boles

Two recent papers analysing nuclear DNA studies and for comparative ecological Australia [12]. These pushed back the sequence data shed new light on the origin and behavioral analyses of the perching known age of the Passeriformes by almost of perching birds (Passeriformes) and the , a model for diversification 25 million years to a time when Australia structure of their radiation. Both papers find comprising over half of all bird species. was still attached to, and just starting to that the New Zealand Acanthisitta Until the 1970s, and throughout the split away from, eastern Antarctica. fall at the base of the passerine radiation, 19th century heyday of description of The next revolution to the traditional implying an origin of this clade in Australia’s avifauna, ornithologists ideas came from the work of Charles Gondwana. Additionally, among oscine believed that the diverse Australian and Sibley and Jon Ahlquist. Based first on (), both papers fail to New Guinea passerines were the poor data from egg-white proteins and then support a sister group relationship between last gasp of radiations that had been on DNA–DNA hybridization [13], the largely Australo–Papuan and accumulating over eons in the Old and they proposed that many of the the Afro–Eurasian , as outlined in New Worlds. Thus, Australo–Papuan Australo–Papuan songbirds (oscine Sibley and Ahlquist’s tapestry. Rather, they ‘robins’, ‘wrens’ and ‘’ were passerines) formed a monophyletic group, converge on a phylogeny in which the thought to have been relatively recent the Corvida, that was sister to the Passerida is nested within the Corvida, derivatives of corresponding groups better remaining groups, which suggesting an origin of songbirds in eastern known and already described from comprised a monophyletic Passerida. This Gondwana (Australia plus New Guinea). Eurasia and the Americas. Such a hypothesis can justifiably be considered Finally, a Cretaceous origin of passerine scenario was consistent with the the greatest achievement of Sibley’s birds is supported by the new data, albeit prevailing paleobotanical view that the 50-year career in avian systematics. It not more on grounds of biogeography than of widespread eucalypt forest of Australia, only implied a autochthonous radiation molecular clocks. The new papers solidify a rather than the extensive tropical of songbirds within Australia that far synthesis of paleontological, phylogenetic rainforest that dominates eastern exceeded in scope that of the marsupials, and molecular data that has been growing Australia and New Guinea today, was but it also suggested that some New and over the past decade, and pave the way for a the ancestral habitat for Old World lineages, such as crows, jays new generation of comparative studies of diversification in this region [9]. and cuckoo-shrikes, ultimately have their passerines. roots in Australia, rather than in the Gondwana rising neo- and paleotropics where they achieve Published online: 06 June 2002 One of the first challenges to the idea of great diversity today. Combined with a northern origin for the Passeriformes recent fossil data, revised taxonomies, The higher level systematics of vertebrate came from paleo-ornithologists. They and paleoenvironmental reconstructions is experiencing a renaissance, noted that, although extensive early implicating rainforest as the habitat and recent data from molecules and Tertiary sites in North America and within which much of this radiation took morphology are revising our Europe contained many small arboreal place [9,14], the DNA hybridization data understanding of relationships within forms, most belonged to derived helped forge a new view of passerine mammals and reptiles [1,2]. Now, nonpasseriform groups, such as the diversification in which Gondwana played two new papers [3,4] bring higher level Coraciiformes, or extinct zygodactylous a leading, rather than a minor, role. relationships of the perching birds groups – none are passerine. Songbirds (Passeriformes) into sharper focus. did not became common in the Northern Phylogenetics, biogeography and Together, they place a final nail in the Hemisphere record until the Middle congruence coffin of the old view that Australia and Miocene, with the earliest unquestioned Both new studies [3,4] employed extensive the rest of eastern Gondwana was the records from Upper Oligocene sites in analysis of nuclear DNA sequence data, recent recipient, rather than the ancient France [10]. The absence of an early primarily because recent analyses of progenitor, of the oscine songbirds, a clade Tertiary record of passerines in the mitochondrial (mt) DNA sequences for comprising today >5000 species. They also Northern Hemisphere led paleo- higher level relationships within birds brighten a phylogenetic spotlight on a ornithologists to propose that the these have been controversial, showing clade of diminutive (and threatened) birds originated in the Southern congruence with few other studies [15]. species in New Zealand, the Hemisphere, before spreading P. Ericson et al. [4] sequenced parts of Acanthisittidae, as the sole survivors of northwards [11]. However, sufficiently the c-myc and RAG1 genes for a total of the sister group to all other passerines. early Southern passerine fossils to test 1428 bp across 21 passerines, whereas These papers, together with a spate of this idea with were not forthcoming until Barker et al. [3] sequenced 3488 bp for recent work on higher level passerine two small fragmentary bones were the c-mos and RAG1 genes across phylogenetics [5–8], now serve as the discovered in an Early Eocene deposit at 69 passerines. Given the similarity of data starting point for future phylogenetic Murgon, southeastern Queensland, sources and methods of analysis, it was

http://tree.trends.com 0169-5347/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S0169-5347(02)02543-0 348 Research Update TRENDS in Ecology & Evolution Vol.17 No.8 August 2002 unsurprising that both papers strongly which hindered clear placement as an across space and time must await recovered a sister group relationship oscine or suboscine. As in earlier work additional southern fossils from more between oscines and suboscines, a result [8,13], this result was moderately strong early Tertiary or even Cretaceous-aged that is in strong agreement with many at best; whereas the Ericson et al. study sites. Although crucial to our morphological and molecular studies. Of could not formally reject trees in which understanding, such new suitably aged greater interest was that both studies Acanthisitta was embedded within the deposits remain elusive. recovered monophyletic New and passerines, the Barker et al. study For neontologists, too, much work Old World suboscine clades, corroborating achieved 80–93% bootstrap values for the remains. Sheldon [19] and others warned the DNA hybridization work [13]. clade comprising all passerines other than of the dangers for comparative biologists Barker et al.’s extensive taxon sampling – Acanthisitta. Noting that New Zealand of investing too much effort in studies including exemplars of 31 of 34 currently was the first to break off from Gondwana relying heavily on poorly resolved or recognized passerine families – is around 82 million years ago (Mya), contentious phylogenies. Given the noteworthy. It enabled them to Ericson et al. suggest that many of the attractiveness of passerines for corroborate one hypothesis for the major divergences within passerines took comparative studies of a variety of relationships of the endemic Malagasy place 71–60 Mya, somewhat earlier than behavioral, ecological and evolutionary family Philepittidae (that it is part of the time frame for family-level radiations traits, the new phylogenies will now serve the broadbills, an Old World suboscine suggested by paleontology [16]. Although as the standard tool for comparative group, Eurylaimidae) and to perform an the sequence data from both studies analyses in this clade. explicit parsimony analysis of ancestral rejected a molecular clock, Ericson et al. References biogeographical areas of passerine were able to employ novel rate-smoothing 1 Murphy, W.J. et al. (2001) Resolution of the families implicating Gondwana as the methods that permit divergence time early placental mammal radiation using cradle of songbird evolution. estimates under these conditions. Still, Bayesian phylogenetics. Science In addition, the two new papers add their dates hinge largely on the single 294, 2348–2351 2 Lee, M.S.Y. (2001) Molecules, morphology and the a significant twist to the emerging story date supplied by the breakup of ancestral of diapsid reptiles. Contr. Zool. of passerine evolution by recovering Gondwana. It would be interesting to 70, 1–22 phylogenies in which Corvida is see what this date implies about rates 3 Barker, F.K. et al. (2002) A phylogenetic to, and paraphyletic with regard to, of nuclear DNA sequence evolution in hypothesis for passerine birds: taxonomic and the Passerida, which is still strongly passerines and, conversely, what the biogeographic implications of an analysis of nuclear DNA sequence data. Proc. R. Soc. Lond. monophyletic. Although phylogenetically consequences of applying mammalian Ser. B 269, 295–308 less tidy than the sister group pattern or other published rates would be for 4 Ericson, P.G.P. et al. (2002) A Gondwanan origin of proposed by Sibley and Ahlquist, the estimating divergences from first passerine birds supported by DNA sequences of new trees favor even more strongly a principles [17]. It seems probable, the endemic New Zealand wrens. Proc. R. Soc. Gondwanan origin for songbirds. At this however, that the passerine radiation is Lond. Ser. B 269, 235–241 5 Cracraft, J. (2001) Avian evolution, Gondwana level, the two papers are striking in their older than was previously thought, a biogeography and the Cretaceous-Tertiary mass congruence, and allow ornithologists result that is consistent with some recent extinction event. Proc. R. Soc. Lond. Ser. B for the first time to implicate eastern mtDNA results. 268, 459–469 Gondwana (Australia and New Guinea) 6 Ericson, P.G.P. et al. (2000) Major divisions in oscines revealed by insertions in the nuclear gene as the birthplace for songbirds. This Digging deeper c-myc: a novel gene in avian phylogenetics. precision was previously not possible However tenuously, a synthesis of Auk 117, 1069–1078 because, as sister groups, Passerida molecular, paleontological and 7 Irestedt, M. et al. (2001) Phylogeny of the major and Corvida had equal importance in phylogenetic results is emerging that lineages of suboscines (Passeriformes) analyzed parsimony reconstructions of ancestral implicates Gondwana as the center not by nuclear DNA sequence data. J. Avian Biol. 32, 15–25 areas, although the Gondwanan only of passerine evolution, but also of the 8 Lovette, I.J. and Bermingham, E. (2000) c-mos distribution of suboscines certainly diversification of the major clades of living variation in songbirds: molecular evolution, favored a similar origin for songbirds [5]. birds in general [5]. There is but a single phylogenetic implications, and comparisons with The two papers are also highly congruent site in Australia (Murgon) in the Early mitochondrial differentiation. Mol. Biol. Evol. in their placement of the taxonomically Cretaceous–Late Oligocene time span 17, 1569–1577 9 Heinsohn, R.G. et al. (1990) Avian cooperative enigmatic lyrebirds Menura spp. and that has produced small avian fossils, and breeding: old hypotheses and new directions. the Australo–Papuan treecreepers the earliest passerine record from South Trends Ecol. Evol. 5, 403–407 Climacteris spp. at the base of the oscine America comes from the Early Miocene of 10 Mourer-Chauviré, C. et al. (1989) Découverte de radiation, albeit with slightly different Argentina [18], and that of Africa even Passeriformes dans l’Oligocène supérior de relationships between these groups. later. Thus, the absence of relevant France. C. R. Acad. Sci. II 309, 843–849 11 Olson, S.L. (1988) Aspects of global avifaunal A final remarkable example of northern fossils at crucial ages finds dynamics during the Cenozoic. In Acta XIX congruence between the two papers is the agreement in the new molecular studies. Congressus Internationalis Ornithologici placement of Acanthisitta, an exemplar of Although several lines of investigation (Ouellet, H., ed.), pp. 2023–2029, National the New Zealand wrens, at the base of the support a southern origin for the Museum of Natural Sciences entire passerine radiation. This group Passeriformes as well as other orders, 12 Boles, W.E. (1995) The world’s oldest songbird. Nature 374, 21–22 possesses primitive states for several this picture remains a broad-brush 13 Sibley C.G. and Ahlquist, J.E. (1990) The morphological traits, including the interpretation. The ability to discern the Phylogeny and Classification of Birds: A Study in morphology of the stapes and syrinx, finer detail of distributional patterns Molecular Evolution, Yale University Press

http://tree.trends.com Research Update TRENDS in Ecology & Evolution Vol.17 No.8 August 2002 349

14 Schodde, R. and Mason, I.J. (1999) The Directory 17 van Tuinen, M. and Hedges, S.B. (2001) of Australian Birds: A Taxonomic and Calibration of avian molecular clocks. Scott V. Edwards* Zoogeographic Atlas of the Biodiversity of Birds Mol. Biol. Evol. 18, 206–213 Burke Museum, University of Washington, in Australia and its Territories, 18 Noriega, J.I. and Chiappe, L.M. (1993) An early Box 353010, Seattle, WA 98195, USA. CSIRO Publishing Miocene passeriform from Argentina. Auk *e-mail: [email protected] 15 Johnson, K.P. (2001) Taxon sampling and the 110, 936–938 phylogenetic position of passeriformes: evidence 19 Sheldon, F. 1997. Phylogeny in studies of bird Walter E. Boles from 916 avian cytochrome b sequences. ecology, behavior, and morphology. In Avian Syst. Biol. 50, 128–136 Molecular Systematics and Evolution Vertebrate Zoology, Australian Museum, 16 Feduccia, A. (1996) The Origin and Evolution of (Mindell, D.P., ed.), pp. 279–301, 6 College Street, Sydney, NSW 2010, Birds, Yale University Press Academic Press Australia.

Gender bias in the refereeing process? Tom Tregenza

Scientists are measured by their Box 1. Citation rates relative to gender, number of authors and country of affiliation publications. Yet anonymous peer review is far from transparent. Does bias lurk within In November 2001, I examined the citation rate of 507 papers published 3–4 years previously the refereeing process? Investigating the in five leading ecology and evolution journals that generally publish authors’ first names, allowing gender to be determined in most cases. Female authors do not appear more likely to outcomes of manuscript submissions use initials rather than their names: of papers where authors used initials, web searches and suggests that the overall process is not correspondence with co-authors revealed that 12 out of 38 were female (2 way χ2 = 2.78, P = 0.1). sexist, but differences in acceptance rates The number of citations for each paper was determined from the ISI citation index. There were χ2 = = across journals according to gender of no differences in citation rate according to gender, (Kruskal Wallis; 3.02, P 0.22) as would be expected if the reviewing process and citation are unbiased in relation to gender. However, the first author give grounds for caution. there were differences in citation rate between papers according to the number of authors, with Manuscripts with more authors and two author papers most cited and a linear decline in citation rate through 3,1 and 4+ authors by native English speakers are more (Kruskal Wallis; χ2 = 8.26, P = 0.041). Hence, although submissions with more than four authors are the most likely to be accepted, they are the least likely to be cited. Country of affiliation of the successful; whether this is due to bias first author has a strong effect on number of citations, even if only wealthy countries are used remains to be seen. (see main text), with papers from non-native English speaking countries attracting fewer citations (Kruskal Wallis; χ2 = 9.17, P = 0.002). Published online: 06 June 2002

Senior academics in the field of ecology the final decision made on their submitted designated as unknown where it was not and evolution are predominantly male, manuscripts over a period of a year. Seven immediately apparent from the author’s in spite of a gender ratio that is closer editors from five journals agreed to take first name (predominantly when only to parity among younger researchers. part providing data from 2680 manuscripts initials were provided) even if the editor Clearly, this pattern is the result of a on the number of authors, the gender of the actually knew the person’s gender. There combination of factors, but a pressing first author, their country affiliation, and does not appear to be any gender difference question is whether gender bias is whether the manuscript was finally in likelihood of choosing to use initials involved. Studies of grant awards [1], accepted or rejected. Gender was rather than first names (Box 1). Data were promotions [2] and resource allocation [3] collected within the period 1997 to 2000. illustrate the potential for sexual Of 2680 manuscripts, 1558 had discrimination in the allocation of 800 male first authors, 498 female, and funding. Potentially equally important, 624 unknown gender, (χ2 = 751, P <0.001). but less scrutinized, is the process by 600 There was no difference in the proportion which decisions on the publication of of manuscripts each gender submitted as 400 manuscripts are made by referees and a sole author (excluding unknown gender journal editors. With the help of the 200 manuscripts) (n = 2056, χ2 = 2.45, P = 0.12). editorial staff of several high impact Number of manuscripts (ISI impact factor >2) ecology and 0 Effect of author gender Unknown Male Female evolution primary research journals, Combining data from all journals (n = 2550 I have examined differences in acceptance Gender of first author excluding those pending a decision), there rate according to gender, nationality and TRENDS in Ecology & Evolution is no overall difference in the acceptance the number of authors on a manuscript. rates of papers according to gender Fig. 1. Acceptance of manuscripts according to gender (χ2 = 4.04, P = 0.13) (Fig. 1). This pattern is The study of first author. There is no overall difference in the the same if only single author papers are I wrote to the editors of 24 primary research acceptance rate of manuscripts according to whether considered (n = 554, χ2 = 1.97, P = 0.37). the first author is designated as male, female or journals requesting that they record the unknown. (Open bars, accepted manuscripts; blue bars, Had there been a difference in gender, nationality, number of authors and rejected manuscripts). acceptance rate relative to gender it would

All in-text references underlinedhttp://tree.trends.com in blue are 0169-5347/02/$ linked to publications – see front matter on ©ResearchGate, 2002 Elsevier Science letting Ltd. Allyou rights access reserved. and PII: read S0169-5347 them (02)02545-4immediately.