Setting the fossil record to rights

Much has been made of ups and downs in the diversity of life from the global fossil record of the Phanerozoic, including the possibility of massive downturns in diversity related to a variety of cause for mass extinction.  However, there are many biases in what is an inevitably imperfect record of biodiversity.  There are anthropogenic influences, for a start.  Although they are becoming more adventurous, palaeontologists cut their teeth on sites close to home, and most of them live in the richer parts of the world.  Insatiable demand for fossils, but mainly of the spectacular and valuable kinds, has grown a world-wide industry of commercial fossil mining.  That may homogenise the geographic coverage of the fossil record, but it is very tempting to go for the richest troves and ignore meagre pickings.  Sedimentation is by no means guaranteed to have been constant through time, partly because of ups and downs of sea level and changes in the pace of erosion of earlier rocks.  Although Phanerozoic stratigraphy seems complete when sections from all over are pieced together, in any one place there are huge gaps of erosion or non-deposition.  It is very easy to come upon several  beds of sedimentary rock and conclude that the sequence represents a continuum in time.  Not so, as any examination of such beds forming today often reveals that intact preservation is the exception compared with erosion and reworking.  The global areas of exposed rocks that cover, say, 10 Ma chunks of Earth history is by no means constant either.  Another factor that conspires to cast doubt on the veracity of the existing fossil record is that the numbers of possible ecological niches that once existed in different tectonic environments are probably not the same.  Active oceanic arcs have few such niches, whereas tropical zones of shallow shelves have vastly more.  There are lots more uncertainties, and New Zealand palaeontologists have painstakingly tried to develop some means of allowing for them in the Tertiary record of their islands (Crampton, J.S. et al. 2003.  Estimating the rock volume bias in paleodiversity studies.  Science, v. 301, p. 358-360).  The simplest premise for estimating bias in the numbers of taxa preserved in rocks covering a particular time range is the available volume of rock from the period that can be sampled.  One approach is to see how geologists have divided up that period in terms of distinct rock formations, the other just uses estimates of the areas underlain by sedimentary rocks laid down during the period.  The first suggests that collecting should be systematically from formation to formation up a sequence, while the second implies that random grid sampling is the best approach.  The New Zealand data suggest that the area approach is most appropriate there, largely because the local rocks formed in a sedimentologically simple, active-margin environment.  Both methods seem to work in tectonically stable areas.  This is just a beginning, but is raises the issue of how much weight can be placed on existing fossil collections in pondering on both titanic and slow-but-sure episodes in the last 544 Ma.

On the same tack, attempts are underway to correct the entire fossil record from 30 thousand collections, using a similar approach to sampling bias.  John Alroy at the University of California, Santa Barbara has helped set up the Paleobiology Database (, following prompting by the most prolific fossil cataloguer, Jack Sepkoski, shortly before his untimely death in 1999.  The web site allows anyone to generate diversity curves, but the process is a little complicated and best tackled by experienced palaeontologists.  You can also enter information from your own collections.  Early results are conflicting.  Sepkoski’s original suggestion that diversity among marine faunas increased since the Triassic may be an artefact of the intensity of sampling which varies from age to age.  However, using just molluscs seems to confirm that at least they did indeed radiate tremendously as Sepkoski had concluded (Schiermeier, Q.  2003.  Setting the record straight.  Nature, v. 424, p. 482-483).

Origins of the vertebrates

Long before techniques were developed to investigate the genetic stuff of living organisms, and when the only known repository of primitive, soft-bodied animals was the Burgess Shale, basic anatomical analysis suggested that maybe the ancestors of vertebrates were worms, sea squirts and even echinoderms.  When the Burgess Shale fauna was re-evaluated and extended in the 1970’s by, among others, Simon Conway Morris of Cambridge University, it became clear that the fossil record was missing a great many delicate and sometimes very odd organisms.  Entirely unsuspected phylla numbered among the occupants of that famous lagerstätte (site of exceptional preservation), but little new about our own ultimate origins.

Vertebrates, echinoderms, sea squirts and a diverse collection of worm-like animals have one thing in common, though apparently very little else.  The first opening to emerge during embryonic development is the anus, whereas in the rest of the animals (protostomes) it becomes a mouth.  So, in the “supergroup” to which we belong, mouths appear at a later developmental stage; hence the sack-name “deuterostome”.  This oddly dichotomous embryonic unfolding points to a very early division among the animals, that might only be unveiled by discovery of even earlier lagerstätten than the Late Cambrian Burgess Shale.  So far, no such source of palaeontological richness has been discovered in late Precambrian sedimentary rocks – crude “molecular clock” approaches to genetic divergence suggest that a great deal went on before the Cambrian Explosion at 544 Ma.  However, the fossil-rich Cambrian of China does push back the record of delicate animals almost to that time.  The recently discovered lagerstätte of Chengjiang is about 530 Ma old, and, as Conway Morris and his Chinese colleagues have discovered, it is rich in fossil deuterostomes.  One group, the vetulicolians, bears a remarkable resemblance to what the pioneer vertebrate palaeontologist, Alfred Romer, suggested as a probable vertebrate ancestor – something with a front end bearing gill slits and a long, segmented tail.  The Chengjiang deposit also contains jawless fish, together with unique “almost fish” called yunnanozoans that may be intermediate links between vetulicolians and fish.  Similarly, there are intriguing hints that vetulicolians evolved towards the most primitive echinoderms, with bilateral symmetry rather than the fivefold form that emerged later.  Clearly, the Chengjiang fauna was extremely diverse and therefore had a long evolutionary history.  Since even more delicate, entirely soft-bodied Ediacaran animals were preserved as imprints in sandstones from the Late Neoproterozoic, it is maybe only a matter of time before low-energy lagerstätten are found from that time.  There are abundant undeformed mudstones from that period throughout the world, but only painstaking rock splitting will find such treasures, unlike the large, “trip-over” Ediacaran trace fossils.

Source:  Conway Morris, S.  2003.  Once we were worms.  New Scientist, 2 August 2003, p. 34-37.


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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s