Attacking convergence

Such multiple evolutionary trials suggest that life is far from some sort of contingent muddle, yet the criticism might be levelled that they are either rather general, such as the example of the matrix of skeleton space, or very specific, good for an anecdote over a pint of beer in the Green Dragon but hardly of any real evolutionary significance. Some examples do indeed seem to be little more than curiosities, such as the Jurassic crustacean that is reminiscent of the trilobites,107 or a fossil bryozoan that adopts a form of spiralling fans that as a bryozoan design had disappeared more than 200 million years earlier.108 Occasionally these reappearances seem more like zombies.109 Such is the case in certain fossil snails found in sediments of Triassic age, that is, the geological interval that followed the great end-Permian catastrophe. They are remarkably similar to the pre-catastrophe snails, but it is not clear whether this is because they are in fact survivors of the debacle (Lazarus taxa) or actually represent new species, but evolving along the same old tracks (Elvis taxa).110

What convergence is, and what it might imply, I hope will become clearer as this book unfolds. A few preliminary remarks, however, may be apposite. First and most obviously, biological convergence can mean many things and operate at many levels. As I shall argue, however, there are some common implications, despite the apparently bewildering range of examples. Second, the branching nature of evolution means that although the identification of convergence presupposes a known starting point, as often as not the reality is an almost infinite regress of investigation. Nevertheless, I believe the topic of convergence to be important for two main reasons. One is widely acknowledged, if as often subject to procrustean procedures of accommodation. It concerns phylogeny, with the obvious circularity of two questions: do we trust our phylogeny and thereby define convergence (which everyone does), or do we trust our characters to be convergent (for whatever reason) and define our phylogeny? As phylogeny depends on characters, the two questions are inseparable. In reality there are, of course, independent data sets and sometimes an historical record, so the enterprise is not crippled. Even so, no phylogeny is free of its convergences, and it is often the case that a biologist believes a phylogeny because in his or her view certain convergences would be too incredible to be true. This is mostly correct, but not always. The second reason for the importance of convergence has certainly been touched upon by others, but, I suggest, is not only neglected but might provide the nucleus of an interesting research programme (see Chapter 10, pp. 308-310). In brief, convergence offers a metaphor as to how evolution navigates the combinatorial immensities of biological 'hyperspace', a topic already touched upon in Chapter 1. Convergence occurs because of 'islands' of stability, analogous to 'attractors' in chaos theory. This view of life, however, begs two interesting questions. First, given the immense dimensions of this 'hyperspace', how is it that navigation is ever successfully achieved? Variants of this question are, of course, repeatedly proposed by opponents of evolution. Convergence gives us a clue as to how the metaphorical Easter Island is located, but just as this Pacific Island is surrounded by oceanic wastes, so too perhaps are the 'islands' of biological habitability. I suspect that not only will the bulk of biological 'space' never be occupied, but it never can be either.

It is therefore my strong belief that the topic of convergence is very far from being just another example of the frustrating imprecisions of biology where apparent rules and laws melt into exceptions and counter-examples. To echo the idea that convergence might give some surprising insights into a deeper structure of biology, there is some evidence that might at first seem to be simply anecdotal. During my time in the libraries I have been particularly struck by the adjectives that accompany descriptions of evolutionary convergence. Words like 'remarkable', 'striking', 'extraordinary', or even 'astonishing' and 'uncanny' are commonplace. It is well appreciated that seldom are the similarities precise,111 and this in itself is as concrete a piece of evidence for the reality of evolution as can be provided. Even so, the frequency of adjectival surprise associated with descriptions of convergence suggests to me that there is almost a feeling of unease in these similarities. Indeed, I strongly suspect that some of these biologists sense the ghost of teleology looking over their shoulders. Nor is this an unworthy sentiment. The eeriness of convergence is central to how evolution navigates across the combinatorial immensities of biological 'hyperspace'. And when we look at some of these examples of convergence, it is not difficult to see why.

Take, for example, the remarkably close correspondence between the first walking leg of the praying mantis and another insect, known as Mantispa.112 As is self-evident, both are modified from a generalized insect leg for grasping (Figure 6.4), yet they belong to distantly related groups within the insects, respectively the mantids and neuropterans.113 The similarity of the raptorial appendage is striking enough, yet there is evidence that such an arrangement evolved independently a third time, in another group of insects, the possibly unfamiliar rhachiberothidids.114 This case exemplifies a more general point because some investigators had earlier allied these insects with the mantispids, whereas a more extensive phylogenetic analysis points to convergence. Each insect therefore seems to have arrived at effectively the same raptorial solution. Note also how in the mantids and mantispids the similarities extend to the large eye and prehensile 'neck'. Incidentally the power, rapidity, and accuracy of the predatory strike, at least in the praying mantids, is quite remarkable.115 The striking action takes two-tenths of a second, and hits home 90% of the time. A standard method of releasing a sudden surge of power is by the elastic release of stored energy, a feature found convergently in many animals and also employed in the catapult. Despite its advantages the mantids do not employ such a system, presumably because of the delays imposed with reloading the elastic. Their remarkable striking ability includes a muscle whose contraction rate equals that of the fastest mammalian muscle.

While on the topic of insect legs it is also worth mentioning that pad-like attachment structures, of which there are two alternative

figure 6.4 A remarkable case of convergent evolution, between the raptorial fore-limb of the praying mantis and the neuropteran Mantispa, both arriving at the same solution independently from the generalized insect leg. (Redrawn from figs. 1-3 of Ulrich (1965; citation is in note 112) with the permission of the author, Natur und Museum and Senckenbergischen Naturforschenden Gesellschaft.)

designs, are also convergent within the flying insects.116 Insects are a rich source of insights into many other convergences, some of which (e.g. compound eyes (Chapter 7), eusociality (Chapter 8), halteres (Chapter 7), silk (see above, p. 115), and trachea (Chapter 10) are mentioned elsewhere. So, too, are the related crustaceans. In the immediate context of attack and defence the repeated evolution of the pectinate claw is a striking example.117 Yet more intriguing is the repeated emergence of a crab-like form118 from among the decapod crustaceans, otherwise familiar in the form of the lobster. This example is of particular interest because, in contrast to many, if not most, examples of convergence where the arrangement has a clear functionality in response to a particular adaptive challenge, in the recurrent emergence (five times) of a crab-like form there is no apparent association with a specific mode of life or environment.119

Let me give you another example, again from the world of attack and mayhem. This is in the form of the independent evolution of dagger-like canines in both placental cats (the sabre-tooth felids)120 and a group of South American marsupials known as the thylacos-milids (Figure 6.5).121 In fact, the evidence suggests that even within the placental felids the sabre-tooth habit evolved several times.122 The dinosaur enthusiast Bob Bakker has also suggested that an analogy to these sabre-toothed mammals can be found in the allosaurids of the Jurassic.123 Although as a group the marsupials, best known in the guise of the kangaroo and wombat, tend to be regarded in some generalized sense as inferior to the placentals, this is too simplistic.124 So, too, the rich, but now largely extinct, diversity of South American marsupials is widely regarded as having been competitively inferior to the onslaught of the placental mammals that surged south when the linkage to North America was secured via the newly emergent Panamanian isthmus several million years ago.125 In fact, in the marsupial thylacosmilids the sabre showed a number of what appear to be design advantages when compared with the placental equivalents, including the possession of a protective flange (the placental sabre-tooth known as Barbourofelis shows the nearest equivalent), a self-sharpening mechanism (presumed to act against some sort of horny pad), and a deeper insertion into the skull that presumably afforded a more secure housing for the canine. It need hardly be stressed that despite this manifest convergence neither group escapes its imprint of phylogenetic history, which is marked, for example, in the specific structure of the teeth.126 It is also worth remarking that the

Marlene Hill Donnelly

figure 6.5 Convergence in the sabre-tooth, between the marsupial thylacosmilid (upper) and placental cat (lower). (Copyright Marlene Hill Donnelly (the Field Museum, Chicago), with permission, reproduced from fig. 10 of L. G. Marshall's The great American interchange - an invasion induced crisis for South American mammals (pp. 133-229), in Biotic crises in ecological and evolutionary time, edited by M. H. Nitecki (Academic Press, Orlando, Florida).)

development of these massive canines, which it appears were more probably used to shear through the flesh of the prey rather than to engage in stiletto-like activities, should not be taken to imply a uniformity of hunting adaptations.127

This instance of sabre-teeth is just one of the many convergences between marsupial and placental mammals. Textbooks tend to emphasize the various similarities between the marsupials of Australia and placentals living in other regions of the world. John Kirsch (note 124) also remarks on the more general convergences between placentals and marsupials, and rightly remarks that such examples are only of real interest if the resemblance turns out to be more than superficial. In particular he draws attention to interesting examples of convergence in the brain structure of marsupials and placentals, notably those involving sensory input through stereoscopic vision and whiskers, while John Johnson draws attention to striking convergences in at least three regional brain specializations,128 examples that are ultimately of considerable significance in the context of the rise of intelligences (Chapter 9).

It is also worth mentioning that a number of striking convergences can be found within the placental mammals themselves,129 such as between the rodents and the group known as the artio-dactyls (hippos, cows, and so forth), as well as among the artiodactyls themselves.130 So, too, broader comparisons between the adaptive radiations of placental mammals that originated in two broad regions, defined as Laurasia (Asia, Europe, and North America) and Africa, define numerous similarities. Thus, in the component supergroups, referred to respectively as the Laurasiatheria and Afrotheria, the many parallels include the emergence of ungulate-like, aquatic, and semi-aquatic ('otter'), insectivore and burrowing, and even ant-eater (note 21) forms.131 Yet more intriguing examples of convergence in mammals come from comparisons between the evolutionary pathways of ungulates and the predators that pursue them. The palaeontologist Bob Bakker132 notes that in six separate evolutionary lines a whole series of anatomical features (e.g. reduction of side toes, elongation of long bones (metapodials) of paws, but shortening of fingers (phalanges)) follow 'the same morphological pathway'. He concludes that 'this striking case of iterative parallelism and convergence ... is a powerful argument that observed long-term changes in the fossil record are the result of directional natural selection, not random walk through genetic drift'.133

figure 6.6 Convergence in the pike morphology, i.e. fusiform, sit-and-wait/stealth fish-hunters, arrived at independently (top to bottom) by the classic pike (Esox lucius, Esocidae: Alaska), Belonesox belizanus (Poecillidae: Central America), Gobiomorus dormitor (Eleotridae: Central America), Acestrorhynchus microlepis (Characidae: South America), Hoplias malabaricus (Erythrinidae: South America), and Hepsetus odoe (Hepsetidae: Africa). (Redrawn from fig. 11D of Winemiller (1991; citation is in note 134) with permission of the author and the Ecological Society of America.)

In their different ways mantids and sabre-tooths are powerful examples of convergence in the context of overpowering prey. Hunting style has, moreover, wider fields of similarity. In an assessment of diversification among freshwater fish Kirk Winemiller drew attention to recurrent patterns of what he called ecomorphological convergences.134 One example is the repeated evolution of an eel-like morphology from phylogenetically divergent sources, as seen in the North American brook lamprey, neotropical swamp eels, and African spiny eel. In the context of hunting, and especially lunging predators, there are again striking convergences on different continents, this time towards a pike-like morphology (Fig. 6.6) and again from ancestors that are only distantly related to each other. Given these convergences among the fish, it is not surprising to see other examples of recurrent evolution. These include similar patterns of teeth, as in sea-breams,135 or related trophic specializations within the endemic cichlids of central Africa.136 Fish, therefore, in their various ways are a rich source of insight into evolutionary convergence, but as we shall see (Chapter 7) there are yet more remarkable examples, such as with the generation of electrical fields and endothermy, that in the wider scheme of things are arguably even more instructive.

figure 6.7 Convergence in desert plants, specifically the stems of the New World (Sonoran Desert) cactus Peniocereus striatus (left) and African (Kenya) spurge Euphorbia cryptospinosa (right). Scale bar is 1 mm. (Reproduced from fig. 3b, g of Felger and Henrickson (1997; citation is in note 137), with permission of the authors and Cactus and Succulent Society of America.)

figure 6.7 Convergence in desert plants, specifically the stems of the New World (Sonoran Desert) cactus Peniocereus striatus (left) and African (Kenya) spurge Euphorbia cryptospinosa (right). Scale bar is 1 mm. (Reproduced from fig. 3b, g of Felger and Henrickson (1997; citation is in note 137), with permission of the authors and Cactus and Succulent Society of America.)

Was this article helpful?

0 0

Post a comment