Biology_of_Turtles
.pdf
56 |
Biology of Turtles |
Nearly all turtles have five vertebral scutes, although anomalies may occur (see examples). Among the few consistent alternatives are the configurations shown by Notochelys platynota (a batagurid) with six or sometimes seven, rarely only five, vertebrals (Figure 3.10), and Lepidochelys olivacea, a cheloniid, in which five to nine vertebrals may be present. The most striking example of costal scute variation is the additional costal at the front of the series on each side in Caretta and Lepidochelys, and even though L. kempii shows only five pairs of costals, L. olivacea may have as many as nine and often shows a different number of costals on each side.
Plastral scute variation is rather limited, except in the families Kinosternidae and Chelydridae (discussed later), but some cryptodires (e.g., cheloniids and dermatemyids), as well as all pleurodires, have an intergular scute. This is small in Podocnemis (a pelomedusid) but may be very large in some chelids (Hydromedusa, Pseudemydura), and in Chelodina it is not only enlarged but recessed inward so that it does not form part of the plastral margin (Figure 3.11). Some systematists use the “plastral formula” as an important character in dichotomous identification keys. This formula ranks the lengths of the midline seams between each pair of plastral scutes, e.g., abd > fem > gul > hum > pect > anal. The formula is reasonably stable within a species but may come unstuck when there is significant sexual dimorphism of the posterior lobe and the anal notch.
Fire, trauma, disease, and so on may cause the scutes of a turtle to separate or peel off. This exposes the underlying bone. The outermost layer of bone, no longer protected by scutes, will necrophy, but the dead bone (or sequestrum) remains in place for a long time, and by the time it finally falls away a new layer of keratin will have formed beneath it. However, the original color, patterning, and texture of the scute is not usually recovered.
Figure 3.11 Anterior of plastron of Hydromedusa tectifera (left) and Chelodina mccordi showing enlarged and distinctive intergular scutes.
Evolution and Structure of the Turtle Shell |
57 |
3.2.3External Form
Turtle shells vary greatly in shape. Some are almost circular (e.g., adult Lepidochelys, Chelodina steindachneri, hatchlings of many species) whereas others may be elongate and narrow (e.g., Chelodina oblonga; Figure 3.12). In lateral profile, the carapace may be highly elevated or domed (e.g.,
Astrochelys yniphora, Geochelone pardalis) or extremely flat (Malacochersus tornieri, Platysternon megacephalum, Apalone mutica). It may be smooth (A. mutica) or rough (e.g., Macrochelys temminckii, Chelus fimbriatus). The carapace may have a single keel (e.g., Pelusios carinatus, Sternotherus carinatus, Graptemys spp., Pangshura spp.), two keels (Platemys platycephala), three keels (many batagurids, Staurotypus spp., many Kinosternon; Figure 3.13), or as many as seven keels (Dermochelys). A pair of transient plastral keels is present in hatchling Podocnemis sextuberculata, and these may also be present in Eretmochelys imbricata.
As a first-order generalization, one might observe that terrestrial species (testudinids, Terrapene, and so on) are inclined to be domed and aquatic ones to be flattened. However, there are numerous exceptions. Moreover, “domed” or “flattened” are not the only two options—for example, a number of riverine species of the genera Graptemys and Pangshura, accustomed to maneuvering in fast currents, have tectiform shells with flat, sloping sides and a high median keel. Some aquatic turtles may have highly elevated shells (east Pacific Lepidochelys olivacea, many Chelodina mccordi) and some terrestrial ones may be very flat (Homopus signatus, Malacochersus tornieri), often as adaptations to life under boulders or in rocky cracks.
Figure 3.12 Extremes within a genus: Chelodina steindachneri (left) and Chelodina oblonga (right), both from southwestern Australia.
58 |
Biology of Turtles |
Figure 3.13 Extreme development of carapace keels in an adult Staurotypus triporcatus.
Some recent studies by Claude et al. (2003) have sought to establish mathematical interpretations of the extent to which the shell morphology of a turtle is shaped by environmental considerations and how much by the phylogenetic history of the species. This procedure requires that the customary analog descriptions of the turtle shell be replaced by a digital system that uses a series of objective “landmarks” on both the bony and the external shell—for example, a typical “landmark” being the point of intersection of a superficial scute boundary or sulcus and a bony suture line. The data are recorded using a three-dimensional computerized recording system (MicroScribe 3D Digitizing System). It was found that both environment and phylogeny play important roles in shaping the shell of a turtle.
3.2.4Kinesis
Once having evolved a rigid all-enveloping corselet, turtles were relatively quick to produce forms in which some degree of flexion or kinesis was possible, although a substantial percentage of living taxa, both aquatic and terrestrial, retain a completely rigid shell. The extreme of kinesis is reached by the small trionychid Dogania subplana (Pritchard, 1993), in which a bizarre condition known as pankinesis occurs with maturity. In this species, the sutural connections between all adjacent carapace bones break down, producing a flat, floppy shell, with an extensive unossified periphery, apparently ideal for finding refuge under submerged rocks and stones in upland streams.
Another dramatic form of kinesis is developed by members of the testudinid genus Kinixys, in which a hinge develops across the rear of the carapace. A hinge across a strongly convex surface would appear to be mechanically untenable, but kinesis is made possible by an infusion of soft tissue between the relevant peripheral bones on each side of the lower edge of the carapace (Figure 3.14).
Evolution and Structure of the Turtle Shell |
59 |
Figure 3.14 Carapace of Kinixys erosa showing posterior hinge.
This tissue is absent along the midline, but flexion at this level is made possible by the presence of a “floating” neural bone that is not tightly sutured to any of its neighbors. The hinge is absent in the hatchlings and young but its potential is suggested by the alignment of the relevant bony sutures and scute seams in the carapace and the unusual, rounded shape of the posterior plastral lobe.
The most common form of shell kinesis occurs in the plastron, and several genera (notably Terrapene and Cuora) have a single transverse hinge across the middle of the plastron that makes possible complete retraction and protection of all the extremities. In other genera (Emys, Cyclemys, Emydoidea, Notochelys, and so on), some degree of transverse mid-plastral kinesis is possible, but its primary function would seem to be something other than simple protection.
Other forms of transverse plastral hinging also seem to contribute little toward physical defense. The different conditions found in kinosternids are discussed later; noteworthy also is the small testudinid Pyxis arachnoides, in which two of the subspecies have an anterior hinge that is not attached to powerful retractor muscles and that moves primarily as a passive result of locomotion or head retraction/extension (Figure 3.15). In a number of turtle genera in which the eggs are few (even just one) but unusually large (Homopus, Rhinoclemmys, Leucocephalon, and so on), there may be some breakdown of the bridge sutures, as well as kinesis of the posterior plastral lobe, to facilitate oviposition.
In the pleurodires, in which the pelvis is reduced to two separated, rigid pillars extending from plastron to carapace, there is little potential for complete shell protection, but the complex genus Pelusios, with over a dozen African species, is distinguished—with the single exception of P. broadleyi (Bour, 1986)—by having anterior plastral kinesis that usually offers very effective protection. In Pelusios gabonensis, the plastron has some degree of longitudinal (midline) kinesis as well as an angled (as opposed to straight) anterior lobe hinge-line (Figure 3.16). Elevation of the anterior lobe is thus accompanied by the slight projection or bulging of the midand hind section of the plastron. Comparably complex modes of kinesis are also shown by certain kinosternids (e.g., K. s. scorpioides), in which the hypo-xiphiplastral sutures meet at an obtuse angle, forcing the sides of
60 |
Biology of Turtles |
Figure 3.15 (left to right) Plastra of Pyxis arachnoides arachnoides, P. a. oblonga, and P. a. brygooi. Only brygooi has no anterior hinge.
Figure 3.16 Angled anterior plastral hinge of Podocnemis gabonensis.
Evolution and Structure of the Turtle Shell |
61 |
the posterior lobe to fold upward slightly when the posterior lobe is raised. This results in exceptionally tight closure of the posterior shell opening.
3.3Sexual Dimorphism
Sexual dimorphism in the chelonian shell manifests itself in several ways, most obviously in the form of a size difference between adult males and female. Within the order Chelonia, some species vary little in adult size, but there are many others in which one sex in much larger than the other. Extreme cases of small males and large females are represented by such species as Graptemys barbouri and Hardella thurjii, where females may have three times the carapace length of males (Figure 3.17). On the other hand, in Galapagos tortoises (Geochelone nigra), the males may reach two or three times the weight of adult females.
In terms of shape, the most familiar sexual shell difference relates to the concave plastron in males of many terrestrial species, a detail probably crucial to successful copulation. There are also frequently differences in the general form of the carapace, which may be relatively rounded or domed in females and elongate and more flattened in males (e.g., Geochelone denticulata, Chelonia mydas). In others (e.g., Testudo boettgeri), the posterior of the carapace may be much wider in males than in females. In saddleback tortoises (Geochelone nigra ssp.), the degree of saddling increases with size, thus reaching a much more extreme expression in the (large) adult males as opposed to the (much smaller) adult females. One also encounters scute softening and de-ossification of the middle
Figure 3.17 Extreme size dimorphism: adult males (top) and adult females (bottom left) of Graptemys barbouri. Center right: Subadult female with fontanels.
62 |
Biology of Turtles |
of the plastral concavity in males of certain unrelated species (Caretta caretta; Phrynops hogei), which may serve as a frictional “assist” for a mating male to maintain his position. In adults of such species as Heosemys spinosa, the adult male has a completely rigid shell with a deep concavity occupying much of the plastral surface, whereas in the female, the concavity is absent and there is extensive kinesis of the posterior lobe (Figure 3.18).
In many turtle species, there may be differences in the shape of the xiphiplastral region of the plastron. This is brought about by the different requirements of males and females as regards the shape of the aperture between the rear of the carapace and the plastron. Females may need to pass impressively large eggs through this aperture, and the space between the xiphiplastra and the pygal bones is rounded accordingly. Males may require a shorter, wider notch to facilitate extension and downcurving of the tail during copulation. In species with extensive plastral mobility, the anal notch is often lacking.
In species with significant size dimorphism, the bony carapace of specimens of small to medium size may reveal the sex of the specimen by the presence or absence of intercostal fontanels. Thus, the carapace of a 10-cm adult male Graptemys barbouri shows no fontanels whereas the carapace of a female of the same size has extensive fontanels. This characteristic reflects the capacity for considerable further growth in the female before maturity is reached, at which point the fontanels eventually close and growth ceases or becomes asymptotic. The reverse is true with Macrochelys temminckii—males 40 to 50 cm in length have open fontanels, whereas these are closed in females of this size, at which they are already sexually mature (Figure 3.19).
In those tortoise species with gular “horns” (single or double), including some Gopherus species, Astrochelys yniphora, Chersina angulata (Figure 3.20), and Geochelone sulcata, the horns are considerably more developed in males than in females, corresponding to the demands of male
Figure 3.18 Adult male (left) and female (right) of Heosemys spinosa.
Evolution and Structure of the Turtle Shell |
63 |
Figure 3.19 Adult female with closed fontanels (right) and subadult male with open fontanels (left) of Macrochelys temminckii.
combat and “vigorous” courtship of females. Nonetheless, somewhat reduced horns are also present in females.
3.4Ankylosis
As turtles approach or reach maturity, the shell fontanels usually close at some point between hatching and the attainment of maturity, but in most species the individual shell bones remain distinct. However, in a minority of species shell bones typically fuse with maturity and the sutures disappear. Such ankylosis commonly occurs in several subspecies of Terrapene carolina (e.g., carolina, mexicana, bauri, but not in most T. c. major), also in some Asian batagurids of the genus Cuora (e.g., members of the C. galbinifrons complex). Some large river turtles (Kachuga kachuga, Dermatemys mawi, Batagur baska, Callagur borneoensis; Figure 3.21), all of which also have very thin shell scutes, also show complete shell ankylosis in adults, and unlike the box turtles this is associated with the disappearance or fusion of the scute seams also. In the smaller, semi-terrestrial batagurid Heosemys spinosa, the shell bones may become fully ankylosed in older adults apart from the transverse plastral hinge in adult females. In other, unrelated species (e.g., Apalone ferox), carapacial ankylosis may occur in exceptionally large, old individuals. Bony sutures generally become more tightly knit and less obvious with ontogeny and onset of old age in many chelonian species (e.g., in Geochelone nigra). In the extinct Mascarene tortoises of the genus Cylindraspis, the shell bones, which were very thin, routinely became ankylosed in adults; generalized ankylosis was also evident in the skulls.
64 |
Biology of Turtles |
(a)
(b)
Figure 3.20 Gular “horn” of male Chersina angulata (left) and female (right)
Evolution and Structure of the Turtle Shell |
65 |
Figure 3.21 Carapace of adult male Batagur baska showing complete ankylosis.
Advantages and disadvantages of shell ankylosis are somewhat speculative—as are the pros and cons of the shell not ankylosing. The condition could be, in part, simply a reflection or by-product of growth slowing and becoming asymptotic or stopping altogether, and certainly full ankylosis of the shell would seem to preclude further growth. In the terrestrial box turtles (Terrapene carolina, and so on), the achievement of shell ankylosis at a rather small size (SCL of less than 15 cm) may correspond to the fact that the moveable shoulder girdle and pelvis of these species may facilitate shell closure but it also severely compromises the weight-bearing capability of these structures (Bramble, 1974). In other words, terrestrial box turtles have to be small or they would not be able to walk, and early shell ankylosis may be the mechanism that ensures that they remain small even after many decades of life. Possibly, this is also the reason that old specimens of Kinixys homeana develop fully ankylosed shells.
The relationship between ankylosis and resistance of the shell to fracture has not been examined adequately, and it is complicated by the fact that some turtles with alkylosed shell bones (e.g., Batagur, Callagur) may indeed have very strong shells, but they also have other features that enhance shell strength, such as heavy buttressing and increased shell thickness. Sutures do not appear to be lines of weakness in living turtles, and examination of traffic-killed specimens of different ages and sizes shows little if any tendency for fracture to occur along sutural lines. Alternatively, Arnold (1979) has suggested that the retention of sutures throughout life may curtail or control the propagation of a crack imposed upon the shell by violent trauma, whereas in a fully ankylosed shell the crack may be unimpeded by sutures and thus spread catastrophically, as in a broken egg.