The Evolution of Modern Penguins and the Importance of Fossils in Molecular Clocks

26 November 2013 by Travis Park, posted in Fossil marine birds, Penguins

The past few decades has seen a revolution in how scientists understand relationships between different species. This is because the methods of sequencing genomes have continually improved as well as the number of different taxa whose genomes have been sequenced has rapidly increased. But for all the amazing new information and data that has come out of all these molecular studies, in order to estimate when one group has split from another you still need the fossils in order to calibrate your phylogeny otherwise you may get results that do not match up with the fossil record. A perfect example of this occurred earlier this month with the publication of a new paper by Sankar Subramanian and colleagues in the Journal Biology Letters.

This particular example concerns those loveable tuxedoed birds of the sea, the penguins. Whilst the oldest known penguins date back to just over 60 million years ago, the timing of the evolution of modern penguins (i.e. the group containing all living species and their most recent common ancestor) has been the subject of considerable debate. A paper published in 2006 by Baker et al. estimated that the crown group (modern penguins) diverged from other penguins around 41 million years ago. The problem with this however is that the oldest fossil penguin that is definitely modern is only around 11 million years old, leaving a 30 million year gap.

Now whilst this gap may well be a real feature, with palaeontologists just not having found fossils of the right age just yet, it has been becoming increasingly apparent that this was most likely not the case. There has been an upsurge in the amount of research on fossil penguins in the past couple of decades, with plenty of fossils being found that date to within that 30 million year gap. Yet not one of them has been found to be from the modern penguin group. Rather, the pattern that emerges is that once the ancient giant penguins disappear in the early Miocene, they are replaced by taxa that are almost modern in appearance, with definitely modern taxa appearing as mentioned above, in the late Miocene.

The phylogeny obtained in the new paper by Subramanian et al. The red 'X' shows the estimate of Baker et al. 2006, the number 4 shows the estimate obtained in this new study. Image modified from Subramanian et al. 2013.

So where does the new paper by Subramanian and colleagues fit into this picture? Well the results they obtained from their analysis suggest that modern penguins first evolved around 20 million years ago, a figure that is much more in line with what the fossils are telling us. Why then did the original paper by Baker et al. get such differing results? A more detailed read of their paper reveals the answer. It turns out that they did use any fossil penguin taxa to calibrate their divergence date estimates, instead using deep external anchor points from other molecular dating studies of other bird groups. These studies therefore may not be appropriate for estimating the dates for more recent events. Furthermore, the Baker et al. (2006) paper only used a single gene in its analysis, therefore restricting its chances of obtaining the correct picture of diversity within the group.

Subramanian and colleagues corrected these issues by using fossils of crown group penguins to calibrate their molecular clock and also used sequences from four different genes, coming up with a picture that is a lot closer in alignment to what we palaeontologists would expect to see. I suppose the moral of this story is, don’t always accept molecular dating estimates on face value. Do a bit of digging and make sure the correct procedures have been followed. And also, whilst those pesky geneticists may think they know it all, they still need the fossils to make their data work, fossils are important! Not that I’m biased of course…



Baker AJ, Pereira SL, Haddrath OP, Edge KA (2006) Multiple gene evidence for expansion of extant penguins out of Antarctica due to global cooling. Proceeding of the Royal Society of London Series B (Biological Sciences) 217: 11–17.

Subramanian S, Beans-Picón G, Swaminathan, SK, Millar CD Lambert DM (2013)  Evidence for a recent origin of penguins. Biology Letters 9, 20130748.

2 Responses to “The Evolution of Modern Penguins and the Importance of Fossils in Molecular Clocks”

  1. Jon Tennant Reply | Permalink

    Cool article Travis.

    This is a problem you actually encounter a lot in molecular research, with respect to using a single gene. What you get with this approach, if you use appropriate modeling techniques, is an estimate of the phylogeny of that gene. However, a lot of the time, these 'gene trees' are conflated with phylogenetic estimates of the organisms themselves, which is just bad science. The best approach is obviously using genomic sequences, calibrated with fossils, as you describe, or at least multiple genes that hopefully converge on the same over-printed phylogenetic signal, or at least resolve to some sort of consensus.

    Yay penguins!

    • Travis Park Reply | Permalink

      Thanks Jon!

      I agree completely, if the authors are not explicit in stating this and present their results and discussion as if it is the phylogeny for the whole species then they are being misleading. Thankfully as the prices of whole genome sequencing continues to fall this issue will become less frequent in publications and we can get as close as possible to the 'real' phylogeny.

      Yay penguins indeed! I'm hoping to have several papers out in the first half next year detailing some Australian fossil material. Work in progress!


Leave a Reply

1 + = ten