An Attempt to Reconstruct the System of Animal Classification

T HE conservative attitude of most zoologists toward any fundamental change in the system of animal classification is to a certain extent understandable in view of the great divergency in such attempts. Moreover, any major change entails obvious practical complications. But every open-minded zoologist is aware that the constant growth of knowledge of the natural history of animals must lead occasionally to a revision of the prevalent system. Important discoveries in the field of general zoology lead to changes in the system of classification. I should like to mention especially the thorough changes in our ideas concerning the relationship between ontogeny and phylogeny or between embryos and ancestors as put forward by de Beer, a subject to which Garstang especially contributed. The interpretation of these relationships has played a great part in the construction of the system of animal-classification ever since the time of Haeckel. At that time, however, too much support was sought from morphology and embryology, while the ecology and physiology of animals were given too little consideration. Thus it happened that primitive characters were confused with characters secondarily simplified either by sessile hkabits or by parasitic life, and that the phenomena of symmetry were incorrectly interpreted. For example, the Radiata were regarded as a phylogenetically uniform group, but it has long been quite certain that radial symmetry has developed in many lines as a consequence of transition from free life to a sessile existence. The Position of Coelenterata