So silken webs and tents full of expectant females may be the order of the day, but wheeled animals steadily progressing across the plains of Africa while admiring a fleet of Fortean bladders seem to be excluded: it appears as if the possibilities of life have been quite thoroughly explored. That we can have some confidence in this conclusion can also be gleaned from various attempts to identify all possible combinations of a particular biological feature. In this instance the approach is to define a matrix that provides a reasonable proxy for combinatorial space, and thus allows one to see whether the same solution to an evolutionary problem has been found repeatedly as against evolving rarely, if not only once. Most interesting in this context are the combinations that 'ought' to exist but do not. A pioneering analysis in this area was by the famous botanist George Ledyard Stebbins, who looked at character combinations in flower structure.73 He found that only about a third of the possible combinations were actually realized, but of the unoccupied majority many were either structurally impossible or evidently poorly adapted. Stebbins also noted that common solutions were due to several factors, including ease of construction and what might be called phylogenetic facilitation; that is, closely related groups of flowering plants are predisposed to arrive at certain floral arrangements. In the latter case it is important to stress that since Stebbins published his paper in 1951 there have been very extensive revisions in plant systematics, especially on the basis of molecular data. It would be an interesting exercise to see in the light of this new evidence how much of the matrix of flower structure is actually occupied by convergent forms.
A similar approach to using a matrix was adopted more recently by Roger Thomas and Wolfgang-Ernst Reif74 in an ambitious attempt to encapsulate all types of animal skeleton, in a so-called 'skeleton space' (Fig. 6.2). This 'space' was defined by such features as shape,
figure 6.2 A plot of 'skeleton space', a matrix that divides the skeleton into seven major categories (and 21 variables), starting with whether they are internal (as in us and other vertebrates) or external (as in arthropods). This gives a total of 186 pair-combinations, of which a handful are effectively impossible. Of the remainder, the majority are either abundant or common, having evolved many times; in rare instances evolution of such an arrangement usually has occurred at least twice. (Redrawn from fig. 12 of Thomas and Reif (1993; citation is in note 74), with the permission of the Society for the Study of Evolution and the authors.)
style of growth and assembly, and degree of interplay. The matrix was defined by a total of 186 possible paired combinations, and significantly it showed that practically no combination, other than the structurally improbable, had not been invented at least once. Indeed, in many cases a position on the matrix showed multiple occupancies: in skeletons convergence is frequent. This leads to a general point, to which I shall return later (Chapter 10) concerning the link between the exploration of biological 'space' (however defined) and convergence, to the effect that barring the physically impossible and adaptation-ally compromised, it appears that as a general rule all evolutionary possibilities in a given 'space' will inevitably be 'discovered'.75
In the context of skeletons it is also worth remarking on some further instances, both specific and general. An interesting example concerns that remarkable biomineral, bone. This shows an immense versatility, particularly in its so-called trabecular form in which thin struts of bone (the trabeculae) orientate themselves according to the prevailing stress field, thereby combining strength with lightness. The precursor of bone, both embryologically, and probably in evolution, is cartilage. In such groups as the sharks and rays cartilage continues to serve as the principal skeleton. Yet clearly this material has design limitations, as is evident from the jaw of the cow-nose ray where a sort of 'trabecular cartilage' has developed. This is structurally and functionally convergent on trabecular bone, presumably evolving in response to the stresses imposed on the jaw as the creature grinds its molluscan prey.76 Bone typifies the vertebrates, and while other groups of animals possess many and wonderfully variable skeletal configurations, it might be thought that bone is a unique biological invention. What could be more different from a vertebrate than a barnacle? These are familiar on seashore rocks and are actually sessile crustaceans, that is, close relatives of the shrimp and lobster. Barnacles encase themselves in a 'box' of calcareous plates, yet remarkably one species (Ibla cumingi) has evolved phosphatic plates with a distinctive and complex structure convergent on lamellate bone.77
Skeletons are usually thought of as tough and resistant, or at least until the snail meets the thrush. Yet biologically the concept of the skeleton can be cast much wider, to soft organs filled with fluid. Probably the most familiar example is penile tumescence when engorged with blood, a useful trick that has evolved several times.78 Moreover, hydrostatic skeletons are an integral arrangement both of entire animals, such as the earthworm, and of specific organs, such as the eye stalks of snails and legs of spiders. In each case the pressure of the effectively incompressible fluid acts as an antagonist to the contraction of a muscular sheath. Hydrostatic skeletons are widespread and in activities such as burrowing often play the key role.79 This is not, however, the case in the various limbless vertebrates, such as the snakes, which, as might be expected, rely on the contraction of longitudinal muscles in association with the backbone. There is, however, a very striking exception in the form of a group of limbless amphibians, relatives of the frogs and newts, which are known as the caecilians.80 It has long been recognized that they are much more efficient burrow-ers than other limbless vertebrates, and it now transpires that this is because they have effectively transformed themselves into worms.81 By allowing the backbone to move independently of the skin, as well as employing a system of helically arranged tendons, the overall style of locomotion is strongly convergent on various invertebrate burrow-ers: a vertebrate body plan has been remarkably transformed into a hydrostatic skeleton.
Was this article helpful?