Growth And Form

Recognizing ancient species_

Paleontologists must interpret fossil species, and their ranges of variation, solely from the morphology, or external shape, of the specimens. There are problems in deciding where one species ends and another begins. When there are close living relatives, it may be possible to compare the modern species with the fossils. But how are paleontologists to decide just what is a species of dinosaur or trilobite?

For modern plants and animals, system-atists ideally apply the biological species concept (see p. 121). Clearly paleontologists cannot test whether fossil species can or cannot interbreed. So, paleontologists use the morphological species concept, judging the bounds of a species entirely on form. The assumption is that all members of a species should look similar, and that a few simple statistical observations should define the mean or average characteristics of members of a v \iX J

VLy Box 6.1 Phenetics and variation within populations

Frequently, paleontologists are faced with problems that require the simplification of a great mass of measurements. For example, a paleontologist may have a large sample of fossils from a single rock horizon and may wish to determine whether these represent one or more species. It might be sufficient to plot univariate frequency histograms (see p. 16) of particular measures, such as width, length and depth of the shells, as well as the hinge width, the diameter of the pedicle foramen, and the length and width of internal muscle scars. In addition, bivariate plots could be prepared, in which various measures are plotted against each other. However, it might still be difficult to differentiate clusters of points, and this approach means the paleontologist has many separate graphs to compare.

Multivariate techniques can help solve these problems by dealing with all the measured variates together. Two common techniques are cluster analysis and principal components analysis (PCA). In PCA, the maximum direction of variation is determined from the table of raw measurements of many characters, and this direction is termed eigenvector 1. Further eigenvectors are then plotted in sequence perpendicular to the first, representing successively less variation in the sample. The first eigenvector usually reflects growth-related or size-dependent variation, and it is typically ignored in taxonomic studies. Species are usually plotted against the second and third eigenvectors, and tests can then be applied to determine whether there are separate clusters of points.

As an example, a comparison may be made between specimens of two species of brachiopod, Dicoelosia biloba from the Early Silurian of Sweden, and D. hibernica from rocks of the same age in Ireland. Four measurements were made on samples of both species and a PCA was performed. Both species were then plotted against the second and third eigenvectors (Fig. 6.1). Although both samples overlap, in general the Irish specimens have lower scores on eigenvector 2, showing that D. hibernica is wider and less deep than D. biloba, and strongly suggesting that there are two species.

Bivariate Analysis Brachiopods

Figure 6.1 Variation in the Early Silurian brachiopod species Dicoelosia biloba from Sweden (o) and D. hibernica from Ireland (+), based upon numerous measurements. A principal components analysis plot separates wide and narrow forms along eigenvector 2, so there may truly be two species, although there is considerable overlap between the two.

Figure 6.1 Variation in the Early Silurian brachiopod species Dicoelosia biloba from Sweden (o) and D. hibernica from Ireland (+), based upon numerous measurements. A principal components analysis plot separates wide and narrow forms along eigenvector 2, so there may truly be two species, although there is considerable overlap between the two.

species and the range of variation around that mean. In practice, this usually seems to be the case (Box 6.1). Critics are correct to point out that the morphological species concept suffers from many problems. Not least is the concern that the true species boundaries might be missed: how do you cope with a single species that has many different forms; how do you recognize species that differ in color, song or smell; how can you distinguish cases where the male and female look very different from each other?

Actually paleontologists need not be too downhearted. Where the biological and morphological species concepts have been applied to particular groups, both generally give the same answers. Further, it would be wrong for a systematist of modern organisms to be too smug. Most decisions on the species bounds of living plants and animals are based on assessments of the morphologies of dead specimens in museums: it is impractical to carry out extensive crossbreeding tests even with living organisms.

Problems with fossil species usually arise from the added dimension of time. If a paleontologist finds a long evolving lineage, where should the dividing line be drawn between one species and the next? Decisions are often made easier by gaps in the fossil record that create artificial divisions within evolving lineages. Where gaps are not present, splitting events clearly mark off new species. If there are few of these, an evolving lineage is divided somewhat arbitrarily into chronospe-cies ("time species"), each being defined by particular morphological features.

Variations in form within species

Within a species, there may be a range of morphologies; think of the variation among humans, or more dramatically, among domestic dogs. In naturally occurring species, morphology may vary as a result of several factors:

1 Individual variation, the normal differences between any pair of individuals of a species that are not identical twins; this base level variation is the stuff of natural selection, as Darwin stressed.

2 Geographic variation and physical differences between populations or subspecies in different parts of the overall species range.

3 Sexual dimorphism, in which males and females may show different sizes, and different specialized features (horns, antlers, tail feathers) often related to sexual selection.

4 Growth stages, where there may be quite different larval and adult stages, or where body form alters during growth.

5 Ecophenotypic effects, where local ecological conditions affect the form of an organism during its lifetime (see p. 123).

Geographic variation may be substantial among members of modern species, particularly those distributed over wide ranges. Sexual dimorphism is seen in living animals, particularly in those where males engage in ritualized displays, or where females have special reproductive activities. Sexual dimorphism is also common in fossils, and it has often caused serious problems of identification where males and females look very different. For example, many ammonites show sexual dimorphism, where the postulated females are much larger than the males, and the males possess unusual lappets on either side of the aperture (Fig. 6.2).

Most organisms change substantially in form as they grow from egg to adult, and these growth stages will be explored next. Ecophenotypic variation was introduced in Chapter 5 (see p. 123) and this includes all the changes in form that may occur through an individual's lifetime, but that are not coded genetically. Ecophenotypic variation in a human might include minor features such as the acquisition of powerful arm muscles through work or exercise or the loss of liver function through alcohol abuse. Major eco-phenotypic changes might include the loss of a limb in an accident or a carefully maintained Mohican haircut. None of these changes can be passed on genetically to a son or daughter by the legless, muscular or unusually coiffed individual.

Allometry_

Changes in form during growth are common. Think of human growth: babies have relatively large heads and eyes, and small limbs. Similar features are found in fossil examples too. Juvenile vertebrates, not just humans, usually have large eyes and heads in proportion to overall body size. A tiny embryo of an ichthyosaur (Fig. 6.3) shows just these features. If measurements of the variable parts (eye diameter, head length) are scaled against a standard measure of the animal (total body length, for example), it is evident that the proportions change as the animal grows older (Fig. 6.4). In the case of the ichthyosaur, the

Figure 6.2 Sexual dimorphism in ammonites, the Jurassic Kosmoceras. The larger shell (a) was probably the female, the smaller (b) the male. (Courtesy of Jim Kennedy and Peter Skelton.)

Figure 6.2 Sexual dimorphism in ammonites, the Jurassic Kosmoceras. The larger shell (a) was probably the female, the smaller (b) the male. (Courtesy of Jim Kennedy and Peter Skelton.)

Figure 6.3 Adult female Ichthyosaurus (a) from the Lower Jurassic of Somerset, England, showing an embryo that has just been born (arrow), and detail of the curled embryo (b). (Courtesy of Makoto Manabe.)

e 10

0 0

Post a comment