Anomalous areas and awkward ages: alleviating concerns
Article first published online: 3 AUG 2011
© 2011 The Authors. Systematic Entomology © 2011 The Royal Entomological Society
Volume 36, Issue 4, pages 604–606, October 2011
How to Cite
WAHLBERG, N. and RUBINOFF, D. (2011), Anomalous areas and awkward ages: alleviating concerns. Systematic Entomology, 36: 604–606. doi: 10.1111/j.1365-3113.2011.00590.x
- Issue published online: 19 SEP 2011
- Article first published online: 3 AUG 2011
- Accepted 15 June 2011, First published online 3 August 2011
Brower & Vane-Wright (2011) critique our recent paper (Wahlberg & Rubinoff, 2011) and bring up some very important points concerning the use of algorithms such as diva (Ronquist, 1997) and beast (Drummond & Rambaut, 2007). Although we agree that the use of algorithms such as these may have some pitfalls and problems, we feel that their judicious use can be informative about the evolutionary history of a group in question. We see the process of delving into a group of organisms as iterative, with each successive study bringing new understanding that can be used to inform the basis of following studies. This is how we have approached our study of the genus Vanessa, a genus belonging to the heavily studied tribe Nymphalini (Janz et al., 2001; Nylin et al., 2001; Wahlberg & Nylin, 2003; Slove & Janz, 2011). We explain below our rationale in light of the critique from our colleagues Brower and Vane-Wright.
We sincerely appreciate the collegial nature of Brower & Vane-Wright (2011), and hope to reciprocate in kind, while also pointing out the difference between the use of a model and blind acceptance of the results. We strongly agree with Brower & Vane-Wright (2011) that the products of a model are only as good as the model and the data, and there are some very real concerns about the function of biogeographical models like diva (Kodandaramaiah, 2010). Although we strongly agree that the results of any model cannot be blindly accepted, we do see value in exploring our data using diva. In Wahlberg & Rubinoff (2011), diva is equivocal about the geographical origins of Vanessa, and includes Hawaii (with its endemic species, V. tamehameha) as a possible origin for the genus. Although diva does not eliminate a Hawaiian origin (among many other regions it cannot eliminate), we never advocate a Hawaiian origin in our discussion, as Brower & Vane-Wright (2011) confirm. The question then, is why use the model at all, as some of the results are so ambiguous (ridiculous). This, we feel, is the important distinction between presenting results and interpreting them. The former should always be done, even if it is to provide fodder for discrediting a model in the future, the latter is the indispensable act of being a scientist. Although we agree that a Hawaiian origin for any part of Vanessa (except the endemic V. tamehameha) is highly unlikely, it is not utterly impossible. If taxa are able to disperse to Hawaii, there is no reason to believe they can never disperse from Hawaii to colonize mainland locations. In fact, there is a precedent for this; Hawaii has been confirmed as the ancestral source region for a globally distributed genus of fly (O’Grady et al., 2008). However, despite this remote possibility, we do not consider that Hawaii is the source for Vanessa and never mention it in our discussion.
The program diva (Ronquist, 1997) is one of the few algorthims available for inferring the biogeographical history of a single clade of organisms. For this reason it is popular, as it is becoming increasingly apparent that each lineage has had a unique history, including a biogeographical history, and dispersal is no longer considered a nuisance parameter, but rather a parameter of interest. However, diva was conceived in the era of vicariance biogeography and it has been shown by several authors to lead to unsatisfactory reconstructions of ancestral distributions (e.g. Kodandaramaiah, 2010). Our use of the algorithm was simply to discover whether there are any robust patterns arising through the biogeographical history of the 22 species of Vanessa. Hence, we used very broad categories for our biogeographical regions and found that there are very few robust patterns in the unconstrained analysis, thus discouraging even finer scale geographical distinctions as unproductive. This is not surprising, given the mobility of the creatures in question and, indeed, this was our interpretation of the results, which we were careful to elaborate on in the text. We chose to show the results of the unconstrained analysis in our fig. 2 (Wahlberg & Rubinoff, 2011: fig. 2) as that is the most conservative representation of the possible ancestral distributions. We specifically discuss the fact that when possible ancestral areas are constrained to five or less, the anomalous reconstruction of Hawaii as an ancestral area of the common ancestor of the genus disappears. We concur that diva is an outdated algorithm and that the results of analyses need to be viewed circumspectly, as we feel we have done. New algorithms are being developed to study the biogeographical history of a clade of organisms that are explicitly model-based and can be used to infer the importance of dispersal. These algorithms show great promise and we look forward to trying them out on our data in the future.
Brower & Vane-Wright (2011) bring up an important point when using calibrations based on fossils. Especially with butterflies, it is often difficult to say whether the fossil represents a sister to an extant taxon or an ancestor of the taxon in question. This problem is compounded by the necessity to constrain a given node in a tree. Simply put, does the fossil tell us the minimum age of the crown group of the organisms of interest? Or does it tell us the minimum age of the divergence of that group from its immediate extant sister group? The condition of the fossil Vanessa is such that it is not possible to place it at any of the previously mentioned nodes with confidence. We thus elected to use information from previous studies (Wahlberg, 2006; Wahlberg et al., 2009) that used the aforementioned fossil as well as other fossils to assign an estimated time of divergence for the first split in the genus Vanessa. Happily, another fossil of the closely related genus Hypanartia (sister genus in some studies, although this position is not entirely stable) from the same Florissant formation is very well preserved and is clearly not related to any of the extant Hypanartia species or species groups (K. Willmott, personal communication in Wahlberg, 2006). Branch lengths of extant Hypanartia species appear to be approximately similar to those of Vanessa (e.g. Wahlberg & Rubinoff, 2011: fig. 1) and, indeed, the age of the first split in Hypanartia is estimated to be similar to that for Vanessa (Wahlberg, 2006). In addition to this, the work by Wahlberg et al. (2009; on which AVZB is an author) used both of these fossils as true minimum constraints on the divergences of the genera from sister lineages. The results suggest that the genera diverged not long before (relatively speaking) the fossilized taxa were extant. Note that the analysis would have allowed those divergences to be much older, but the data suggested otherwise. Taking all this evidence together, we feel confident that Vanessa is not much older than our analyses suggest in Wahlberg & Rubinoff (2011).
We point out that we included the very broad error parameters for our age estimates and discuss the ages in the broadest of terms, which makes a distinction of 5 million years well within the boundaries of our discussion. Molecular clocks must be interpreted as broad suggestions, especially when based on equivocal fossil taxa (Graur & Martin, 2004). Thus, although our error estimates suggest 25–31 million years, the very process of molecular dating has always necessitated an intellectual interpretation that is broader than the statistical one. As with the preceding issue of biogeography, presenting the data and interpreting it effectively are separate issues. Brower & Vane-Wright (2011) rightfully draw attention to the importance of the distinction. We agree and feel that our recent work, while presenting controversial results, does not cross the line into advocating scenarios supported by models but not common sense. We concede that others may have different views on the age of butterflies and Lepidoptera in general, and welcome alternative interpretations through data analysis, thus hopefully improving our understanding of the evolutionary history of the fascinating creatures we enjoy studying.
We thank Andy Brower and Dick Vane-Wright for a healthy discussion, some valid and interesting points and giving us a chance to put one more paper in our CVs. Our work was supported by the Academy of Finland (grant number 118369) to N.W. and by Hawaii Agricultural Experiment Station Hatch Funds (HAW00956-H) to D.R.
- 2011) Anomalous areas and awkward ages: concerns about over-reliance on model-based biogeographical and temporal inferences. Systematic Entomology, DOI: 10.1111/j.1365-3113.2011.00590.x. & (
- 2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology, 7, 214. & (
- 2004) Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision. Trends in Genetics, 20, 80–86. & (
- 2001) Evolutionary dynamics of host plant specialization: a case study of the tribe Nymphalini. Evolution, 55, 783–796. , & (
- 2010) Use of dispersal-vicariance analysis in biogeography – a critique. Journal of Biogeography, 37, 3–11. (
- 2001) Phylogeny of Polygonia, Nymphalis and related butterflies (Lepidoptera: Nymphalidae): a total-evidence analysis. Zoological Journal of the Linnean Society, 132, 441–468. , , , , & (
- 2008) Out of Hawaii: the origin and biogeography of the genus Scaptomyza (Diptera: Drosophilidae). Biology Letters, 4, 195–199. & (
- 1997) Dispersal-vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology, 46, 195–203. (
- 2011) The relationship between diet breadth and geographic range size in the butterfly subfamily Nymphalinae – a study of global scale. PLoS ONE, 6, e16057. & (
- 2006) That awkward age for butterflies: insights from the age of the butterfly subfamily Nymphalinae (Lepidoptera: Nymphalidae). Systematic Biology, 55, 703–714. (
- 2003) Morphology versus molecules: resolution of the positions of Nymphalis, Polygonia and related genera (Lepidoptera: Nymphalidae). Cladistics, 19, 213–223. & (
- 2011) Vagility across Vanessa (Lepidoptera: Nymphalidae): mobility in butterfly species does not inhibit the formation and persistence of isolated sister taxa. Systematic Entomology, 36, 362–370. & (
- 2009) Nymphalid butterflies diversify following near demise at the cretaceous/tertiary boundary. Proceedings of the Royal Society of London B, 276, 4295–4302. , , , , , & (