Read an Excerpt
Animal Species and Their Evolution
By A. J. Cain PRINCETON UNIVERSITY PRESS
Copyright © 1993 Princeton University Press
All rights reserved.
ISBN: 978-0-691-02098-3
CHAPTER 1
INTRODUCTION
The science of biological classification is called taxonomy, and it is the task of the animal taxonomist to describe all the known forms of animals, to sort out their relationships so that they can be classified in as natural a way as possible, and to provide a system of nomenclature, so that each form can be referred to rapidly and accurately. In recent years it has been gradually realized that taxonomy is not merely a necessary pigeon-holing but also one of the most important activities in biology, requiring a synthesis of all other biological pursuits for its proper performance, and producing results of the highest importance in the study of evolution. That good classification is essential for all zoological work is obvious. No one would think much of a chemist who confused water and benzene "because they look alike". His classification would be defective, and might lead to serious practical consequences. Similarly, the equally incompetent zoologist who thought that whales were fish would soon find on further investigation that the apparent resemblance is misleading, and that in most features they agree with mammals. Or again, the forester who applies the remedies against a particular pest to his trees should be sure that it is that pest he is dealing with, and the physiologist investigating the 'common earthworm' will not confirm the results of others unless he is awake to the fact that there are several sorts of common earthworm, each with its peculiarities.
It is the business of the zoologist to study animals. But there are vast numbers of animals in the world, and they present an amazing diversity of structure, habits, and mode of life, as anyone can see if he will examine carefully a crayfish, starfish, jellyfish and dogfish. And if he will look at all the starfish, for example, that he can lay hands on, he will find there are many different kinds of starfish. Probably most people can tell apart several closely related sorts of bird or butterfly. But few know that there are forty sorts of earthworm in Britain alone, and probably several that, for lack of investigators, have never been described. Over half a million different forms of insect have already been described, and more are discovered almost daily. Even in Britain, one of the best-worked countries in the world, many kinds of animal remain to be discovered, and the fauna of huge areas in the tropics is hardly known at all.
Unfortunately, because of the difficulty of sorting this vast array of diverse forms, it has not infrequently happened that the taxonomist, studying a particular group and flooded out with demands for identification, has been unable to go further than naming his specimens and describing them in just sufficient detail to permit them to be distinguished from their nearest relatives. Often he has had to work entirely on preserved material which he did not collect himself and with very insufficient field-notes attached. And since a great deal of his time was taken up with finding the right name for a particular specimen, often changing well-known names in the process, there grew up almost a tradition among research workers in other fields of zoology that the taxonomist was purely a 'museum man' engaged in sorting skins or shrunken pickled specimens of animals he had never seen alive, using diagnostic characters that no one else could see, and wholly taken up with endless futile controversies about names. There was some little truth in this caricature. Even the mild and charitable Darwin said in a private letter, "... I have long thought that too much systematic work [and] description somehow blunts the faculties," although he added that the particular taxonomists he was writing about were "a very good set of men."
In the last twenty years there has been a gradual revolution. It is generally admitted by taxonomists on the one hand that to describe any particular sort of animal adequately one must take into consideration not merely its structure and distribution but also its genetics, mode of life, physiology, and all its other aspects. On the other hand geneticists, physiologists and students of evolution have realized the importance of good taxonomy, and in studying whole groups of animals at a time have been able to make generalizations of the utmost importance. In particular, the origin of species (a subject hardly touched upon by Darwin himself) is now far more clearly understood and is recognized as a most important stage in evolution. The old reproach that no one had ever seen a species evolving into others is no longer valid. And a comparison of the process of evolution as seen in action today with the fossil records of it in the past strongly suggests that the same sort of process acted then as now, and was as responsible for the main evolutionary lines as for the side-branches.
Until very recently, the species was thought of as merely one rank in a classification which was meant to bring together those animals that most resemble each other in small groups that could themselves be grouped into larger ones and so on upwards until one came to the group of all animals. In modern taxonomy several meanings of the word species can be distinguished, of which the most important is that it is the lowest group of animals which at least potentially form an interbreeding array of populations, unable to interbreed freely with other sorts of animals. It represents the attainment, in Dobzhansky's words, of "that stage of the evolutionary process at which the once actually or potentially interbreeding array of forms becomes segregated into two or more separate arrays which are physiologically incapable of interbreeding." The process of speciation is therefore of the utmost importance in evolution. Without speciation, specialization and the development of great efficiency in the exploitation of particular modes of living would hardly be possible, since the acquisition of special adaptations by some living organisms in one part of the earth would be continually hampered by outbreeding to others in adjacent regions. As climate, soils and waters vary all over the earth, the effects of selection to withstand particular conditions in one district would be nullified by constant interbreeding with organisms that had been subject to selection for withstanding the opposite conditions.
In this book, the older and simpler meaning of the species as merely one rank in the natural classification is first discussed (Chapter I-IV). Its extension, to take into account the extraordinary geographical variability of species (Chapter V) that was discovered when zoological exploration was begun on a large scale, was the beginning of the realization that structural differences are not always safe guides to species-limits, and that criteria of interbreeding in the wild are of greater importance. This realization has led up to the so-called biological definition of the species (Chapter VI.) This definition cannot be applied to all animals, and the treatment of asexually reproducing forms, and of 'species' that are portions of a single evolutionary lineage, requires special discussion (Chapter VII). Recognition of the importance of geographical variation, and the formulation of the biological definition of the species make it possible to show that probably the most important process in speciation is the production of geographically isolated populations which can become genetically so different while they are isolated that when they meet again they cannot interbreed freely. This is the theory of geographical speciation (Chapter VIII). And finally, mechanisms of speciation are considered (Chapter IX) that do not require geographical isolation, of which one certain example is known, and more may be discovered.
CHAPTER 2
METHODS OF CLASSIFICATION
There are many different ways of classifying a set of things. We might arrange them according to whether or not they possess some particular character, and then subdivide the groups so formed by the presence or absence of another character, repeating this procedure until all the things in each of our smallest groups were for our purposes identical. This method would give a hierarchy of groups, each of which contained two others with mutually exclusive characteristics. For example, one could divide all living things into those that fly actively and those that do not. The group of fliers could then be divided into those with one pair of wings and those with two (since no animal with three is known). Those with one pair could be divided into those that have hair (the bats) and those that have no true hair (birds, many insects), and so on. The virtue of such a classification as this is that if the characters used for diagnosis are selected for being conspicuous, it will be extremely easy to work through the classification and find out to what group any particular specimen belongs. Such classifications are therefore used for identifying and are usually called 'keys'. In the example just given every group contains two others. It is a dichotomous key. Sometimes a group may be divided into three or more according to convenience. One could divide the group of fliers with one pair of wings into those with hair (bats), those with feathers (birds), and those with jointed bodies (insects). But in general, the dichotomous key is used, as being the simplest and easiest to work through.
Alternatively, one could remove from one's set of objects all those with some striking peculiarity into one group, those with another into another and so on, until one had a large number of mutually exclusive groups of equal status, instead of a hierarchy. The difficulty in this procedure is that one may have to read through descriptions of all the groups to identify one specimen, and that the last group is usually a rag-bag full of all those objects without obvious distinguishing characters. In the insects, the group of Neuroptera, until it was restricted to the lacewing flies and their allies was just such an assortment, and the old classification, so well known to everyone, of the Lepidoptera into butterflies and moths distinguished one fairly clear-cut group (the butterflies) while leaving together a large array of others (the moths) each of which is now correctly accorded at least equal status with butterflies. Another difficulty is that, particularly in sets containing large numbers of closely related forms, it is hard to find a series of diagnostic characters of which all specimens have one and one only. For example, one might group the British butterflies into those having some white somewhere on the wings, those with some yellow, those with some blue, etc., but some have both white and yellow, or both yellow and blue. The difficulty is intensified when the males and females are remarkably different. One way is to prescribe the order in which the characters are used, and that transforms this method into a hierarchical method straight away. For example, one could divide butterflies first into those with some and those with no white, and then each of those groups into those with some and those with no yellow. Often, the taxonomist who has studied a particular group well can say on looking through a key that it was clearly constructed by removing all the obvious subgroups first, and was then licked into a dichotomous shape. The symptom is the rag-bag group at the end. Rag-bag groups always give trouble except in keys, which are made solely for convenience in identifying. They are defined solely by what they are not, not by what they are, and nearly always they turn out to be very miscellaneous groups.
The two methods just given produce mutually exclusive groups, whence their utility for keys. Either a specimen has the distinguishing character of a certain group or not, and if not, it must go into some other. But they rely at each division on the use of a single character. Unfortunately, otherwise unrelated animals may have characters in common. Both bats and butterflies have wings, but bats are vertebrates, butterflies are insects. They differ fundamentally in all other characters. And with objects as complex as animals it very frequently happens that in one or two members of a group which is otherwise very clearly defined, that very character which one would like to use to diagnose the group is absent, although all the other characters of the organism concerned leave no doubt that it is a member of the group. For example, the group of birds (warm-blooded vertebrates, with feathers, and laying large yolky eggs) is a very natural one, and the possession of a pair of wings for flight is one of its most characteristic features. One can recognize a flying bird as a mere silhouette a mile off, when its possession of a backbone, feathers, warm blood and the capacity for laying large yolky eggs are not readily discernible. But the kiwi, for example, cannot fly. Yet it is certainly a bird. Here, one can remove the difficulty by simply changing the diagnostic character, and using not flight but feathers. Again, everyone is accustomed to the idea that mammals are warm-blooded vertebrates with true hair (like we have ourselves), that bring forth their young alive, and suckle them by means of mammary glands functionally developed only in the female. In the echidna or spiny ant-eater, and the duck-billed platypus, the hair and mammae are present, but an egg is laid. In this respect and in many features of the skeleton, they are closer to some extinct reptiles. However, on the whole, their features are mammalian, and they are accordingly grouped with the hedgehog, the bat, the whale, the cow, the cat and ourselves. The difficulty here is rather more acute, as their inclusion makes the mammals a more heterogeneous, less easily defined group. In some series of fossils it becomes intolerable. In many series of fossil mammals, although the latest forms are very distinct and specialized, more generalized forms are found in older rocks, and finally one gets down to a mass of very closely related forms. In each series leading from this mass to the end-forms, there may be so profound a change that no single character remains unaltered and available as a diagnostic, except those that are too broad for use within the group. For example, in the great group of primates (tree-shrews, lemurs, monkeys, apes and man) it is almost impossible to find a single diagnostic character (running through the whole group) which is not so basic that it says only that all are mammals, and therefore fails to divide them from hedgehogs and other insectivores.
In short, reliance on a single character will not only group together unrelated forms (as when in our first example we lumped bats, birds and flying insects merely because they all fly) but may even get us into a position where we can produce no diagnoses at all, and have no one character to rely on, although the group we are dealing with is obviously a good one—that is, its members have more in common one with another than they do with members of other groups. It is essential, if we wish to group similar animals together, to take account of all features, and look for general resemblances rather than particular differences.
This changeover from classification by difference to classification by resemblance is of the utmost importance. The distinction between the two methods was clearly understood by some earlier naturalists, especially the great taxonomists John Ray (1627–1705) and Linnaeus (1707–1778). Indeed Linnaeus, while establishing a classification of plants employing effectively diagnoses by single characters, and providing keys so good that they are still used in schools in Sweden, explicitly sets out the idea that in the most natural classification (not the one most easy to use for identification) all possible characters must be taken into account, no single one being used intolerantly. This 'best' system may be referred to as the natural system of classification, all others being called artificial. In the natural system, animals are grouped according to their basic similarities into as many groups and subgroups as their resemblances and differences require.
(Continues...)
Excerpted from Animal Species and Their Evolution by A. J. Cain. Copyright © 1993 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
