The Succession of Faunas

Thus it appears that evolutionary change in animals may be either directed towards an improvement in structure which adapts the animal for some particular mode of life, or it may take place without any apparent relationship to habits. Evolutionary change which is not adaptive in nature may be expected to take the same course in related lineages, and adaptive changes will be similar in such of their members as have the same habits and are sub jected to the influence of like environments. (Further examples of this phenomenon will be found in the articles HORSE; PERISSO DACTYLA ; MAMMALIA ; and REPTILES.) Adaptive Radiation.—Another phenomenon first emphasised by H. F. Osborn, is that which is known as "adaptive radiation." The early members of any group, the Basal Eocene placental mammals for example, are small, and on the whole very uniform in structure; as time goes on their descendants radiate so as to fill all habitats open to them. Thus the mammals are fitted for life in forests, plains, deserts and mountains, for flight or for an aquatic life in rivers, along the sea coast or in the open oceans. They may eat animals or plants, the prey may be small or large, and it can be caught either by chase or by stalking. They may live in the Arctic or in the tropics. In consonance with these varied habits the structure of mammals has become exceedingly diverse. The limbs may be long, freely moveable and with grasping hands and feet ; as in arboreal animals such as lemurs. They may become reduced to flippers in the whales ; the digits may end in hoofs among cursorial animals, or may be clawed in carnivorous ones. The dentition may be fitted for gnawing, for cropping grass or for tearing flesh from bones, and it is usually possible to deter mine from its character what were the habits of the animal.

The "adaptive radiation" of all the higher groups of vertebrates follows similar lines. Reptiles during Mesozoic times filled the places now occupied by mammals and sometimes present striking similarities to them in general appearance.

A

survey of the evolutionary history of animals as a whole shows that for each group there is a certain period at which the beginnings of an adaptive radiation appear and within a very short time lead to the establishment of the great majority of the families of which it is composed. Thus the ammonoids, although they had been in existence since Devonian times, are represented only by two families in the Carboniferous. In the Permian and Trias they branch out into most diverse forms, clearly fitted for very different modes of life. Teleost fishes exist ing from Upper Triassic times are represented only by a single family in the Lower Cretaceous. By the Upper Cretaceous nearly a dozen families have appeared and by Eocene times the majority of those which are known were in existence. Reptiles, beginning in the Carboniferous, experienced a wide radiation in Permian times, whilst the "radiations" of birds and mammals lie in the Tertiary, and were already fully established at the end of the Eocene. Thus in all these cases the initiation of the main types

took place together and led very rapidly to the establishment of separate families, which then undergo a steady evolution without bringing about fundamental alterations in their structure. This fact implies that only at its origin and from forms of conserva tive structure can a group give rise to new lines whose members differ from their ancestors in fundamental features. In other words the possibility of change becomes more and more restricted the further the members of a group have carried a process of adap tive radiation. Nevertheless certain of the smaller groups of all classes may exhibit an adaptive radiation of their own long after the main development of the class is completed.

Migration.

The fact that the evolution of non-adaptive features pursues a parallel course in related lineages, and proceeds little interfered with by modifications of the environment, at a rate which does not vary widely, provides us with a means of determining approximately the age of rocks which may not con tain known or even fully identifiable fossils. For example any rocks anywhere in the world which contain members of the high est grade of labyrinthodonts, will be of Triassic age and even from previously unknown forms it may be possible to determine whether they belong to the upper or lower part of that system.

The use of individual species of fossils to determine the iden tity of horizons in widely separated parts of the world suffers from a theoretical objection which was pointed out by T. H. Huxley, who suggested that an individual species probably came into existence at some definite place and that in order to reach any other locality its individuals had to migrate. Nothing was known of the time taken in such migrations but it was conceivable that it might be comparable with such small divisions of geo logical time as those represented by zones. He stated that rocks containing identical faunas were homotaxial but not necessarily contemporaneous. We now know chiefly from the evidence of fossil mammals that all the larger groups have an evolutionary home where the important part of their evolutionary change takes place, and that from this centre there is a constant migration in all directions of such a nature that the series of forms in some neighbouring area, although they fall into a morphological family tree, are not in fact directly descended from one another. Thus all evidence shows that the main evolution of the horses took place in northern Asia and that the series which is found through out Tertiary times in North America is really built up from the members of many overlapping migrations, each new colonist giving rise to a short evolutionary series of its own. On the other hand the camels arose in North America and the forms there found build up a true phylogenetic series.