October 3, 2016:

by Tony Edger

In which the blogger persists in his mad quest to find connections between two disparate subjects, come hell or high water.
I began my summer studying literature on the evolution of the turtle shell and ended it engrossed in a wonderful doctoral thesis on British theater playbills. Unexpectedly, I found that these two topics had something important in common. Both speak to the challenge of identifying and understanding transitions. Recognizing where, when, why, and how transitions occur, say between one species and a daughter species, or from one function for a specific body part to another function, is a core issue in paleontology, one often particularly difficult to determine from, or map in, the fossil record. As to British theater playbills, the doctoral thesis I read grappled with the transition of playbill to play program, offering, in the process, a perspective on a core issue for understanding transitions – it’s critically a matter of definitions. [Note: This discussion below of the turtle shell draws heavily on an article I wrote for a fossil club newsletter.]

Turtles

In an article published in July this year, paleontologist Tyler R. Lyson and his colleagues provocatively hypothesized that the turtle’s shell got its start in the Permian (some 260 million years ago) in the lizard-like reptile, Eunotosaurus africanus. (Fossorial Origin of the Turtle Shell, Current Biology, July 14, 2016) To understand the nuances of Lyson’ hypothesis, it’s helpful to first consider the shell in extant turtles. The shell consists of two principal elements: the arched carapace on the dorsal side of the animal, and the flat shield-like plastron on its ventral side. Pictured below is the skeleton of a yellow-bellied slider (Trachemys scripta scripta) on display at the Smithsonian’s National Museum of Natural History. A section of the carapace has been cut and hinged to expose the plastron. The horny scutes, that, in life, adhere to the bones that make up the carapace and plastron, have been wired to the skeleton. (The signage at the Museum is silent on whether this is an actual skeleton or a cast.)

Overall, Lyson’s is a striking hypothesis, challenging the widely held belief that the turtle shell evolved as a defensive structure, and that its initial stages occurred in an aquatic environment. The entire issue is awash in questions of transitions, of where to draw lines. Most critically, Lyson’s hypothesis rests on an identification of so-called “stem turtles,” those species that we consider antecedents to “crown turtles,” as well as his characterization of the environments within which the known stem turtles lived. A stem turtle shares some, but not all, of the traits we deem necessary for a species to be considered a true or crown turtle.

Is Eunotosaurus actually a stem turtle as Lyson would have it? The challenge is determining where the transition from non-turtle reptile to stem turtle occurs, and there are consequences for how we do that. If Eunotosaurus is the earliest known stem turtle, then, not only might the fossorial hypothesis about the shell’s original purpose be correct, but so would terrestrial and Permian beginnings for turtles.

But, if it isn’t a stem turtle, then there are younger candidates among known fossils which appear to tell a somewhat different story. Among these are the following two.

1) Pappochelys rosinea, dating from the middle Triassic (about 240 million years ago), had, in addition to what appears to be an incipient carapace, paired gastralia (bones that some reptiles have in their abdominal walls) and some ossification and fusion in the ventral region which, while not constituting a plastron, seem headed in that direction. The animal lived in a lacustrine (i.e., having lakes) environment. Paleontologists Rainer Schoch and Hans-Dieter Sues, who first described Pappochelys, conclude that it probably “lived along the lakeshore or frequently entered the lake,” and that its features were “consistent with aquatic or semi-aquatic habits” (A Middle Triassic Stem-Turtle and the Evolution of the Turtle Body Plan, Nature, Volume 523, July 30, 2015, p. 587.)

2) Odontochelys semitestacea, which lived about 220 million years ago in the Triassic, was, until recently, very widely accepted as the earliest known stem turtle. Although it had an only partially evolved carapace with some broadening of the dorsal ribs, its plastron was fully developed. (Chun Li, et al., An Ancestral Turtle from the Late Triassic of Southwestern China, Nature, Volume 456, November 27, 2008.) Fossils of this animal have been found in what was, in the Triassic, a near-shore marine environment, which lead Li et al. to conclude that the characteristics of Odontochelys are “indicative of primarily aquatic habits and of a possible aquatic origin of turtles.” (p. 500.)

I am struck by another question that poses somewhat of a transition issue. How should we characterize a lacustrine environment? Lyson looked at the available candidates for stem turtles and concluded that Odontochelys was the outlier because of the fairly clear aquatic nature of its place of origin. He argued that the rest, including those coming from lacustrine environments such as Pappochelys and the stem turtle with the earliest complete turtle shell in the fossil record, Proganochelys quenstedti (Triassic, about 215 million years ago), should be considered terrestrial in origin. I do wonder whether environments which feature lakes are to be so easily deemed terrestrial. Isn’t it possible that the species found in such environments lived in the lake waters, or moved between lake and land in their quotidian activities?

Also, I see an argument to be made that the morphological changes that paleontologists see as leading to carapaces and plastrons might have had two concurrent functions – aid for burrowing and for defense. An animal that spent even just part of its time in lake waters might benefit from, at least, the beginnings of a plastron to deal with predators coming from below, and might also do some burrowing when on land, aided by an incipient carapace. Further, did these ancient species hibernate, a burrowing activity that modern turtles living on land and in water may engage in?

There might well have been other functions of the various features of the turtle shell. For instance, some have suggested that the plastron served as ballast for aquatic species, helping them stay properly oriented.

Odontochelys offers up yet another transition conundrum. Why is Odontochelys’s plastron the earliest completed element of the turtle shell that we know from the fossil record? It's already been suggested that a water dwelling species might benefit from the protection it provided against predators coming from below or for its possible function as ballast. But, there are alternative hypotheses about where in the development of the turtle Odontochelys should be placed. Paleontologists Robert R. Reisz and Jason Head, for example, suggest that Odontochelys was part of a radiation of turtles from land to sea and that the partial carapace was the remnant of what had been, in some older species, a complete carapace. (Turtle Origins Out to Sea, Nature, Volume 456, November 27, 2008.) In other words, this species was in the process of losing the carapace while retaining the full plastron which was functionally useful in a marine environment. Further, this suggests there may be fossils from terrestrial stem turtles older than Odontochelys with complete shells waiting to be discovered, and that evolution of the plastron and carapace occurred in an essentially terrestrial environment, possibly in support of burrowing.

I have to assume that finding more fossils that further illuminate the evolutionary process that led to today’s turtle will also fuel continuing debates about whether some particular new fossil find represents a stem turtle or not, and what flows from that.

Playbills

Late this summer, when I acquired two antique British playbills, I found myself, once again, exploring transitions.

Some context for this acquisition is probably in order. When I was a college student, I worked for two summers as a page at the Folger Shakespeare Library in Washington, D.C., retrieving items from the closed stacks and vaults of the library for researchers. An ongoing project of mine was to assign dates to myriad 18th century British playbills, many of which omitted the year of the performance. Memories of my time spent with those thin, often yellowed sheets of paper with their relatively crude printing flooded back when, in August, I was wandering through an antique store and came upon two very early 19th century British playbills in frames. Their allure was irresistible; they now hang over my desk. (They are roughly 7 1/2 inches wide and 12 1/2 inches high.)

Perhaps the most telling assertion Gowen makes is that there is no precise, absolute way to classify playbills as to their function, content and form. From the outset, apparently, playbills defied easy characterization. The roles they assumed, the formats they displayed, the text they contained varied over time. It wasn’t some linear trajectory. At times, early in their history, they were simply announcements of an upcoming performance, but soon they also became more, advertisements for a play, for actors, and for theaters. They often concurrently served as aids to understanding and following a play when they began providing such things as a listing of the cast and a precis of the plot. They sometimes grew to a large size, the better to promote a performance, but theater patrons were known to fold them to make them more manageable. They were ephemera, destined to disappear (covered over, torn down, weathered away) once a performance took place, even as they were also long preserved mementos, ways of remembering a performance. They were pasted to posts (yes, there were posts erected for this purpose) and walls while also being distributed as loose flyers. They were distributed for free and also sold. On occasion they were spread around town for free in the morning and sold at the theater in the evening.

So, what, at first blush, might seem an easy distinction to make – between playbill and program – is not. Without universally accepted agreements on the structure of a playbill, what information it conveyed, and how it was to be used, it may not be surprising that drawing a clear line in this instance may be, according to Gowen, impossible to do. For instance, one might argue that, when the playbill assumed the function of supporting an audience member in attendance at a performance with appropriate content, a transition had been made and the playbill was now effectively a program. But, Gowen writes,
If, based solely on its content, the programme is seen to have evolved from the playbill by 1737 with the regular inclusion of a cast list, a more formalistic approach towards the classification of this principal variant of the bill of the play suggests that the programme emerged more than one hundred years later, in the 1850s, by which time the content of playbills averaging approximately twenty-six inches in height and seventeen inches in width was reformatted and printed in small format impositions. (p.172)
So, in the 18th century, even if the function and content of the playbill were sometimes serving the theater patrons during a performance (arguably hallmarks of a program), the form of the playbill had yet to make the relevant transition. Interestingly enough, I don’t see Gowen arguing here that the transformation of playbill to program occurred in the 1850s. Rather, his position seems to be that, given the whole muddled history of the British playbill and program, this distinction lies largely in the eye of the beholder, and perhaps it’s not a distinction that can be made.

The lessons learned from reading Gowen’s thesis offer some help in thinking about questions of transitions in paleontology. On the one hand, the notion that much depends upon one’s definitions is critical and relevant (what attributes must an ancient species have to be considered a stem turtle? is there universal agreement on that or is it up for debate?). On the other hand, throwing up one’s hands in surrender (as Gowen does), because no one else has fashioned widely accepted definitions of the parameters to be applied, isn’t really helpful.

Nevertheless, I cannot fault him too severely because, as far as doctoral theses go, this one is a pip. Readable, accessible, informative, and often funny (e.g., there was a code of conduct among the folks who pasted up playbills which, among other things, discouraged covering over some rival’s playbill if the glue on it wasn’t dry).

There’s one final instance of a problematic transition that emerges from these playbills of mine. Both announce a performance by Edmund Kean in the lead role of King Lear at London’s Theatre Royal, Drury Lane – the first for August 30, 1820; the second for February 24, 1823. (One passing observation – in that 1820 performance, Junius Brutus Booth played Edgar, the legitimate son of the Earl of Gloucester. Booth was the father of John Wilkes Booth, Abraham Lincoln’s assassin.)

The question of a transition centers on the play itself. In his compelling The Year of Lear: Shakespeare in 1606 (2015), English professor James Shapiro describes how, as early as the First Folio, efforts were made to lighten the darkness, the brutality of the play. As a result, what was acted for some 150 years as Shakespeare’s King Lear was, he asserts, actually Nahum Tate’s revision of the play in which “Lear lives and Cordelia and Edgar will marry and inherit his kingdom.” David Bevington and David Scott Kastan assert that “the appeal of his [Tatum’s] sentimentalized adaptation was so powerful that Shakespeare’s play simply disappeared from the theater for a century and a half.” (William Shakespeare King Lear, edited by Bevington and Kastan, 2004, p. xxiii.)

So, when is King Lear actually Shakespeare’s King Lear? What can be missing without threatening that identity? Are the darkness and brutality of the ending essential to his play? Are Lear’s death and Cordelia’s death critical? Is the Fool crucial?

My playbill of 1823 actually marks part of an important reversal, offering a seminal point in the transition back to a play that is increasingly the one that Shakespeare wrote, rather than being, at its core, Tate’s. In a performance on February 10, 1823, at the Drury Lane Theatre, a great deal of the play was restored to its original form. This was considered, by some, to be “the first fully recorded performance of the play given approximately as Shakespeare intended it to be acted.” (Oscar Fay Adams and A. Wilson Verity, Introduction, in The Works of William Shakespeare, edited by Henry Irving and Frank A. Marshall, Volume VI, 1889, p. 331.) Though it did not go far enough for other critics. Still more of Shakespeare was returned to the play in the performance announced by my bill dated February 24, which states in brackets below the title of the play, “With original Passages of Shakespeare restored.” Nevertheless, is this version really Shakespeare’s Lear when the Fool still hasn’t made it back and Cordelia and Edgar continue to be part of a love story? I guess that depends upon your definition of what constitutes King Lear.