Thus for each character there is a story of an incoming of a definite evolutionary stage in a small percentage of the indi viduals living at a certain time. As we pass into newer rocks the percentage displaying this stage increases until the vast majority of all the specimens found conform to it, and we then have living a few individuals in an evolutionary stage which was dominant at an earlier period, and a few precocious specimens which have a structure which will later become the common form of the major ity. In the case of Micraster the change in structure is perfectly gradual, every conceivable intermediate between the two extremes occurring in one specimen or another. Thus on the evidence of one character alone it is possible to say that a given Micraster must have been found in one of three zones and that there is a probability, which could be expressed as a percentage, that it was from the middle of the three zones. It is the belief of all those who have studied the distribution of Micraster in the field that it is possible to determine their horizon with much greater accuracy by examining not only one but all of the characters which Dr. Rowe has shown to change in a similar manner. This possibility depends on the fact that an individual Micraster which exhibits a slower evolution than usual of one character, will present an acceleration of development for other characters, and that re garded as a whole it will be equivalent in stage, though not identical in structure, with those that are found with it. One character, the ratio of breadth to length of the whole test, does not exhibit a parallel evolutionary change; broad and narrow forms are present in each zone in about the same proportions, so that it is clear that this character is not undergoing a steady evolution. The broad and narrow forms at any one time exhibit a similar range of variation with respect to any one character, and it is perhaps reasonable to assume that the broad forms are descendants of one another and are not mere varieties of the associated narrow species. That this is indeed so is indicated by the fact that in the stage of evolution of their ambulacra they always lag behind the narrow forms.
The general features of the evolution of Micraster as set out above are repeated in the evolution of all other series of animals which are well-known. In each case change of structure of any organ or character proceeds in a definite direction and appears to change gradually and not by definite steps. In every case an advanced structure is first met with in a small proportion of the individuals living at a definite time and this proportion grows larger and larger as we proceed upwards. Finally most individuals possess it but a few retain a more primitive structure, whilst the rest are more advanced, foretelling the character of the majority in a succeeding period. An animal may be regarded as built up from many quasi-independent characters each exhibiting its own series of evolutionary changes ; and characters of any one indi vidual will be found to differ in the evolutionary stage which they have reached, so that by striking an average of the evolutionary stages of all the characters, it is possible to determine the date of the individual with considerable accuracy.
Members of closely allied but distinct groups such as those represented by the broad and the narrow Micrasters, undergo a parallel evolution, but the rate of evolutionary change may differ when one group is compared with another. A great part of mod ern palaeontological work consists in the attempt to sort out the members of such closely allied groups from one another and the establishment of series of forms known as phyla or lineages, which at any one time contain only a single species. This process has perhaps been carried farthest in the case of certain Cretaceous Bryozoa where W. D. Lang has distinguished a very large number of lineages each of which exhibits an evolution parallel to, though differing in detail from, that of the others. The study of lineages presents many difficulties because in the case of animals with a simple skeleton there are not many structures which, persist ing unchanged throughout the history of the lineage, enable us to distinguish it from all others. Under these circumstances it is always possible to regard the members of the presumptive lineages as mere varieties which have arisen at each period from a main stock in which the whole of the evolution is taking place. Mod ern work on genetics has shown that identical mutants are con stantly arising from normal animals and that the mutants of allied species and even genera take the same forms. This evidence lends support to the view that the alternative explanation may in many cases be preferable to the accepted distribution of allied forms amongst lineages.