Tuesday, December 29, 2020

Finding Meaning in a Bunch of Fossil Teeth

Sharks are cartilaginous fish and, typically, very little of them fossilizes, except for their teeth.  An individual shark loses and replaces teeth throughout its life, in the process shedding thousands or tens of thousands of teeth.  No wonder the typical collector may end up with seemingly innumerable fossil teeth stored (sometimes randomly) in baggies, jars, drawers, and display mounts.  Studies of extant sharks show that rates of loss and replacement may slow with age and are affected by a host of other variables, including species and feeding behavior (Visaggi and Godfrey 2010; Purdy 1998; full citations of all sources provided at the end of this post).   Making some meaning out of those myriad teeth is a challenge for any fossil hunter.  A challenge I failed to meet when I once had a great opportunity.

A decade ago, I wrote a post describing an attempt to replicate the findings of a published analysis of two large collections of Miocene fossil shark teeth from the Calvert Cliffs (Visaggi and Godfrey 2010).  Paleontologists Christy Visaggi and Stephen J. Godfrey analyzed the distribution by genus of two groups of teeth that had been donated to the Calvert Marine Museum by collectors:  nearly 1,900 teeth collected in situ as they were exposed in the cliffs, and somewhat more than 24,000 teeth collected from the wash on the beach, so-called float teeth.  The authors identified a number of biases that influenced either of these two sources of teeth including:  payoff bias in which collectors went to where teeth were most likely to be found, and size and prize bias in which collectors skipped over small teeth and gathered up primarily large, rare, or unusual specimens.  (For the latter bias, the really fine specimens were unlikely to be donated and so were mostly missing from the group of teeth found in the shoreline wash.)  The sheer number of specimens and the two different sources of the teeth were believed to reduce the influence of these collection biases.

I attempted to reproduce the results by spending one day collecting every shark tooth I could find in the wash (thereby attempting to address the size and prize bias) along a specific, productive stretch of beach under the Calvert Cliffs (of course, ensuring that my teeth reflected the payoff bias).  I ended up with 157 teeth.  Pictured below are a few of the specimens I collected on a very cold March day in 2010.

The distribution by genera of my sample mirrored rather closely that of the float teeth analyzed by Visaggi and Godfrey for their localities labeled CC9 and CC10 (see Visaggi and Godfrey 2010).  The most common genera represented in their float collection for these localities were:

Carcharhinus (Gray Sharks – 51% of their float teeth versus 45% of the teeth I collected)

Carcharias (Sand Tiger Sharks – 9% versus 8%)

Galeocerdo (Tiger Sharks – 18% versus 17%)

Hemipristis (Snaggletooth Sharks – 9% versus 7%)

Isurus (Mako Sharks – 3% versus 3%).

The most prominent genera in my collection also included these two others:

Negaprion (Lemon Sharks – <1% versus 4%)

Sphyrna (Hammerhead Sharks – <1% versus 3%)

(The teeth from these last two genera are small and so, not unexpectedly, are largely missing from Visaggi and Godfrey’s float portion.)

That’s as far as I went with my analysis.  It seemed sufficient to generally confirm with my one-day effort and 157 teeth what Visaggi and Godfrey found by rigorously analyzing thousands of teeth.  I failed to draw any deeper meaning from my results or those published by Visaggi and Godfrey.  Though they drew attention to what I think is a fascinating aspect of the results of these efforts, I missed it entirely. 

Visaggi and Godfrey stressed that their results applied only to the relative distribution of teeth from different shark taxa, and did not speak to the relative abundance of individual sharks from those taxa.

We do not wish to imply that the percentage of shark teeth in either the float or in situ collections are an accurate reflection of the actual numbers of individual sharks that lived in Salisbury Embayment during the Miocene (Visaggi and Godfrey 2010, p. 31; the Salisbury Embayment was the geological precursor to today’s Chesapeake Bay).

Teeth not individuals.

But . . . not withstanding that statement, I think they do leave the reader of their paper with the implication that the number of teeth are probably not divorced from the numerical diversity of the populations of sharks in those Miocene waters, particularly if the focus is on higher taxonomic groupings:  i.e., orders of sharks.  (Orders are the fourth taxonomic grouping counting down from kingdom, genera are the sixth.)  I should clarify:  an attentive reader would have recognized the implication.

Let me explore that further.  When the seven genera listed above were grouped into their orders, what I failed to see ten years ago became painfully obvious.  Only two orders of sharks are represented.  Two of these genera – Carcharias and Isurus – belong to the Lamniformes order (mackerel sharks) while the remaining five belong to the Carcharhiniformes (ground sharks).  Significantly, as is clear from the percentages given above, the vast majority of teeth collected along the cliffs comes from Carcharhiniformes sharks (over three-quarters of the teeth).  

A shark from each of these two orders is shown below in drawings from Kent (2018).

Sand Tigers are an example of the Lamniformes.

Grays are an example of Carcharhiniformes.

Ichthyologist Leonard Compagno and colleagues have described extant representatives of the former (the Lamniformes) as “mainly large active pelagic sharks” (Compagno 2005, p. 175), and the latter (the Carcharhiniformes) as sharks that are mostly “small and harmless to people, but this order also includes some of the largest predatory sharks” (p. 186).

Significantly, the relative shares of teeth accounted for by each of these orders is the opening to an interesting story of what was happening taxonomically to sharks during the Miocene:  Carcharhiniformes were coming into their own, ultimately surpassing the Lamniformes.  It’s a story I am only now exploring.  In 2010, Visaggi and Godfrey wrote:

Lamniforms flourished from the Cretaceous through the Eocene, whereas carcharhiniforms remained less common.  The diversification of carcharhiniform sharks occurred during much of the Paleogene and by the Neogene carcharhiniforms had surpassed lamniform sharks in abundance.  Both data sets [in situ and float collections] demonstrate the dominance (>70%) of carcharhiniforms . . . over lamniform . . . and other sharks (<30%) during the Miocene at least in terms of teeth produced.  This preeminence of carcharhiniform sharks persists in modern marine environments.  (Godfrey 2018, p. 32, emphasis added)

Paleontologist Bretton Kent’s recent guide to the Miocene fossil shark teeth of the Calvert Cliffs is instructive in this regard (Kent 2018).  He described teeth from 8 genera and 16 species of Lamniformes, and teeth from 9 genera and 16 species of Carcharhiniformes.  (As delineated above, this relatively equal balance in genera and species between the two orders is misleading because it masks the distribution of teeth that a collector along the Calvert Cliffs shoreline will come upon.  My March 2010 collection and the Visaggi and Godfrey paper show as much.  And, as noted, there is an important implication possibly to be derived from those numbers of teeth,)

When Kent analyzed changes in the morphology of shark teeth by individual species from the Cretaceous to modern times, he found that teeth featuring tall, narrow crowns and cusplets (small side crowns) diminished in prevalence by species and were replaced by teeth with a more diverse mixture of morphologies, including the increased presence of serrations.  For example, as is shown by the following pictures of two teeth from the ones I collected in 2010, Carcharias teeth exhibit the former morphology (first picture below), while Carcharhinus teeth show the latter.

Kent posited that teeth with narrow crowns straddled by small cusplets enabled their possessors to grab and restrain prey, while those with serrations supported cutting and clutching prey.  Significantly, he then wrote:

By the late Oligocene and early Miocene shark faunas have a more complex mixture of tooth morphologies comparable to that of the Holocene.  This change is partially correlated with a fundamental shift from faunas dominated by lamniforms to one dominated by carcharhiniforms (Kent 2018, p. 47).

There, that’s the story for which my collection of teeth a decade ago was an early chapter:  one order of sharks in the process of displacing another as the dominant one.

That change is even more robustly reflected in today’s oceans.  Carcharhiniformes are clearly the lead order.  Compagno and his colleagues have identified only 15 extant species of sharks belonging to Lamniformes, while Carcharhiniformes include 225 living species.  The latter order, they wrote, “is the largest, most diverse and widespread group of sharks” (Compagno 2005).

What of the Chesapeake Bay today compared to the Miocene Salisbury Embayment?  Is this changeover reflected there?  Comprehensive, rigorous data describing the shark taxa presently found in the Bay are somewhat elusive.  I certainly have no estimates of the actual number of sharks by species in these waters.  The best data I have found come from the Chesapeake Bay Program (CBF), a partnership of federal and state agencies, nonprofits, and academic institutions working for the restoration of the Bay.  These data suggest that the dominance of Carcharhiniformes worldwide is somewhat reflected in the Bay.  The CBF identifies 12 species that are rare to common in the Bay and adds an additional 4 species for which there is a single record of a sighting in modern times (Eney 2010).  Of these 16 species, Carcharhiniformes account for 11 while Lamniformes are represented by only 2.  But, for both orders, these counts of species are only somewhat marginally related to the relative abundance of these sharks.  Given the decimation of shark ranks in the modern era from human activities, apparently only two of the Carcharhiniformes species are considered common in the Chesapeake Bay as is only one of the Lamniformes species.   

A decade ago, I was swept up in the effort to replicate the Visaggi and Godfrey analysis, and failed to understand the story revealed by what I was finding.  This reminds me of a line from T.S. Eliot’s Four Quartets:

We had the experience but missed the meaning.


Leonard Compagno et al.  2005.  Sharks of the World.

Lindsay Eney.  2010.  Are There Sharks in the Chesapeake Bay?  Chesapeake Bay Program.  August 4.

Stephen J. Godfrey, editor.  2018.  The Geology and Vertebrate Paleontology of  Calvert Cliffs, Maryland, USA.  Smithsonian Contributions to Paleobiology Number 100.

Bretton W. Kent.  2018.  The Cartilaginous Fishes (Chimaeras, Sharks, and Rays) of Calvert Cliffs, Maryland, USA.  Chapter 2 in Godfrey, The Geology and Vertebrate Paleontology of Calvert Cliffs, Maryland, USA.

Robert Purdy.  1998.  Fossil Shark Teeth.  The Paleontological Society.

Christy C. Visaggi and Stephen J. Godfrey.  2010.  Variation in Composition and Abundance of Miocene Shark Teeth From Calvert Cliffs, Maryland.  Journal of Vertebrate Paleontology.  January.

Nature Blog Network