Developmental trait evolution in Trilobites

Fusco et al. Evolution 66(2): 314-329. DOI:10.1111/j.1558-5646.2011.01447.x. Developmental trait evolution in Trilobites

This is a guest post with Tim Astrop. Below, we give our first impressions of this article. Please comment below, or tweet TimWill or Lynsey (maybe use #pegejc). Think of this as a journal club discussion group!


Figure 1 from Fusco et al. – trilobite moult cycle

Tim Astrop

Tim Astrop

This paper is a wonderful example of how innovative thinking and an integrative approach to the fossil record continues to discover new keys with which to unlock the unknown biological information held within. The Trilobitomorpha is arguably the most diverse and abundant extinct arthropod subphylum, there is certainly no shortage of palaeontological research on these critters, that’s for sure. But despite such popularity and interest in the group we still know very little about them as biological organisms. In this respect it is very apparent that preserved remains are not representative of the living animal, there is much interpretation to be done (see the recent ‘spotted trilobite’ described by McRobets et al 2013).

By looking at different ontogenetic stages represented by the different instars (discrete developmental stage of arthropods ‘bookended’ by ecdysis, or molting) preserved for over 60 species of trilobite for which adequate data is available, the authors then derive new metrics to treat growth rate via average per molt growth increment (AGI) and conformity to Dyers rule (IDC), while the former metric is self explanatory, the latter is less intuitive. The IDC quantifies the fit of observed ontogeny to that of a constant growth rate, deviations from Dyers rule would be indicative of accelerated/decelerated growth at particular ontogenetic stages (readers with a thing for shape analyses may also have noted that allometry would upset this metric via modular differentiation of growth rates within an organism, well done, have a gold star).

What is particularly cool is that the authors use this info not just to look at the evolution of this group but also to shed light on early euarthropod development, as the Trilobitamorpha are basal arthropods. Euarthropod evolution is a very active area of research (see Yang et al 2013 for interesting recent discovery) and held by many as something of a rosetta stone for understanding arthropod origins and diversity.

The meat of this article has a little too much math for than I would normally elect to tackle, but after making it through the results the discussion highlighted the findings really well.

In a nutshell, it seems that trilobites conform to Dyers rule overall, with some smaller deviations occurring in early (protapsid) development. However, the particularly interesting thing, for me at least, is that there appears to be less phenotypic integration in later stages that enables very unique, differing morphologies to evolve that deviate from standard cut-and-paste metamerism (segment repetition) seen in many trilobites (but probably most evident in the familiar Myriapoda) and calls forward to the future of arthropod diversity in all it’s mis-matched-crazy-appendage glory. It also infers that that similar morphologies may be converged upon via different developmental pathways, to me this indicates that trilobite morphology is not restricted to an adaptive peak but provides a method to move between peeks via morphological lability, something the authors refer to as ‘evolutionary dynamism’, a trait that likely contributed to their success as a group.

In summary, I think this article is a wonderful example of how valuable palaeobiological approaches can be to understanding organismal evolution. Here we have a novel approach to an existing dataset using linear measurements (and complex stats) to derive new information about the evolution and biology of an extinct arthropod subphylum.

In my opinion, the true integration of biological and palaeontological principles allows fossils to be treated as the cryptic remains of biological entities that old-school paleontology can often overlook in favour of a ‘stamp-collecting’ mindset. Thankfully there appears to be something of movement among palaeobiologists at the moment back to a more synthesized ‘Simpsonian’ concept of the fossil record. Something I look forward to hearing more about, and to hopefully contribute to!

Will Pearse

Will Pearse

I normally hate people who loudly shout “yay science” on Facebook and the like, but this paper really was one of those moments for me. We can track the development of growing trilobites using fossils? We can robustly test hypotheses about the evolution of that development? Yay science!

The most parsimonious evolutionary model of these species’ traits is one where phylogeny doesn’t matter (a star phylogeny), so it’s hard to argue that there’s any constraint to trait evolution here, despite the phylogenetic signal of some of these traits. The authors suggest this reflects the influence of particular outlier clades that are biasing model fit; I completely agree. I spend much of my time doing analyses of ‘plants’, and it’s always struck me as rather strange that I lump all these diverse and divergent species under one label and then examine their evolution altogether. I imagine the same is true of trilobites – the devil is in the detail, and understanding exactly what clades are varying (and why) will probably give us a better handle on what drives the evolution of these traits.

In passing, I wonder what the consequences of doing morphological analyses on phylogenies of extinct species that have to be built from morphological data. What if some of these traits, or traits that are strongly coupled with these traits, were used to define the phylogeny of these species – after all, ‘ontology recapitulates phylogeny’, and information about how species develop would seem (to a person ignorant of paleontology) like useful data. Could this be why the authors find stronger phylogenetic signal in the earlier developmental stages, or does this actually reflect the biology of the system? Does more evolutionary constrain in juvenile growth suggest juveniles were competing more strongly than adults (mortality does tend to decrease as species get older)? The authors give very detailed supplementary information on the phylogeny’s construction, but I’m really nowhere near qualified to comment. Can anyone help me out?

Lynsey McInnes

Lynsey McInnes

You can’t deny that Will and I like to stretch ourselves in the diversity of papers that we discuss on this site. Who knew the coverage of the terms phylogeny, ecology, geography and evolution was so broad. This week we take another foray into paleo research, this time without the potential to incorporate any extant data, these things are extinct!

My first reading of this paper was definitely made while wearing a cynical hat, and I was unconvinced that the data were adequate to ask these questions, particularly regarding the patched together phylogeny. A bit more time with the paper though, and I was becoming more and more impressed with the authors’ ability to extract conclusions, all appropriately caveated given the antiquity and patchiness of the available data. Perhaps we are all becoming too obsessed with perfect datasets, Bayesianed up and with no room for interpretation based on slotting in genuinely new findings with an already amassed body of research. This paper is a great example of the latter approach. Neat!

I’m most acquainted with adaptive zones in terms of diversity dependent cladogenesis…clades diversify until a zone is full, diversification rate declines until some upstart clade pops off into a new zone to start the process again. Sadly, a lot of contemporary ‘adaptive zone’ research remains a numbers game with little detailed investigation in what adaptations a clade possesses to occupy their zone. In contrast, this paper makes a concerted attempt to actually think about the adaptations necessary to occupy different zones and finds strong evidence for these traits in the Trilobite lineages studied. Impressive given these things went extinct millions of years ago! Not only that but the authors also successfully distinguish between intrinsic and extrinsic drivers of the patterns that they find! I’m a bit obsessed with the landscape or environmental drivers of macroevolutionary patterns and was really pleased to see this angle tackled in this paper.

So, I’m no paleontologist or developmental biologist, and appreciated this paper mostly as it reminded me that science need not be all about the perfect dataset producing good model fit in a standalone clincal model, it can be messy and incomplete, but illuminating.


About will.pearse
Ecology / evolutionary biologist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: