to the intensity of the stresses to be withstood. So far well; but the further step is to utilize the principles of physics in accounting, as far as may be, for the vital architecture. A guillemot lays a single top-shaped egg on a narrow shelf of a precipitous sea-cliff. Now Darwin pointed out that when an egg of this shape is roughly jostled by the wind or by the parent bird, it will rapidly rotate on its short axis, but will not roll away from the spot. The mechanical adjustment is simple and effective, but the further question is how far the adaptive result can be accounted for in terms of the physical conditions of pressure, surface-tension and the like, operative on the egg-shell while it is still in the making in the oviduct. It is for the student of biophysics to work this out.
Enough is not yet known of the possibilities of another line of physiological morphology, represented by the work of Prof. C. M. Child, Individuality in Organisms (1915), which seeks to interpret structural differences in terms of different rates of metabolism at the various levels or " gradients " of the axis of body-symmetry.
Homoplastic Structures. Criticism of the concept of homol- ogy (resemblance in embryonic origin and fundamental structure) has brought into prominence the kind of resemblance which Sir Ray Lankester called " homoplasy, " to which others have applied other terms, such as " convergence." It is a resemblance in types not closely related, and is due to the similar adaptation of non-homologous structures to similar conditions of life. But it is plain that the idea requires careful handling. There are cases where homologous parts become adaptively very like one another, directly or indirectly in relation to similar function- ing, as in the case of the " paddles " of ichthyosaurs, turtles, penguins and cetaceans. To such cases it has been proposed to apply the term "parallelism," while Prof. Gadow (1913) has carried the analysis further, distinguishing " isotely " (hitting the same mark), e.g. zygodactyl feet of cuckoos, parrots, and woodpeckers, from " homaeotely " (hitting the same target, but not quite the same mark), e.g. the jumping foot of kangaroo, jerboa and Tarsius, Both of these have to be clearly separated from what many call true convergence, which Prof. Gadow has called " parately," where the feature has been evolved from parts and material so different that there can be little relationship, if any. The resulting resemblance is more or less superficial. Thus the bivalve shell of Brachiopods has nothing to do with that of Lamellibranchs, and the bulla auris of certain extinct reptiles (Pythonomorpha) is probably the quadrate, while that of the cetacea is the tympanic. Similarly, it is probable that the eye was evolved many separate times. Now the well- marked convergence or homoplasy of non-homologous struc- tures depends on a cooperation of factors: e.g. (a) the wide- spread occurrence of certain similar hereditary factors which will respond much in the same way to similar environmental conditions; (6) the plastic power of similar functioning, operating in individual development; and (c) the similarity in the sifting process of natural selection. Along with (a) may be included the idea that the number of new departures is probably far from being unlimited; along with (c) may be included the idea that the biophysical conditions of stability are within relatively narrow limits ; and along with (b) the similar moulding of similar functions may be included, as Gadow says, the subtle effects of correlation. It is probable that the continuation of the interesting study of convergence, along the lines indicated by Lankester, and more fully by Willey and Gadow, will lead to a deeper understand- ing of the principles of organic architecture or morphogenesis. It may also throw some light on classification, for it is worth considering whether the same congeries of characters may not arise more than once, and that not fortuitously. Prof. Gadow refers to his finding in Mexico the burrowing snake, Typhlops braminus, previously known from islands and countries of the Indian Ocean basin, a fact which suggests to him that the species had evolved twice. Some would say indeed that three similar domestic dogs might be traced to three different ancestors, a coyote, a jackal, and a wolf. The question also arises whether this promising study of convergence or homoplasy will not
also elucidate close pattern-resemblances between forms, in regard to which it is impossible to suggest any genuine mimicry.
The Question of Species. A huge number of new species of animals are described annually, sometimes several thousand insects in one year. When a collection has to be reported on or a new region explored, there is obvious need for the old-fash- ioned definition of new species, i.e. forms which do not conform to any previously described. To give one of these novelties a new name, if it is marked off by peculiarities of some magnitude (necessarily to some extent a matter of opinion), is practically a much more desirable procedure than the restraint which says: " Mus n. sp., in the vicinity of Mus syhaticus," for this makes subsequent reference intolerably tedious. But several changes may be detected in the zoological species-making of to-day, (i) There is no longer much enthusiasm for this line of activity in itself. It is a necessity, a useful discipline; but there is no particular hurry in completing the descriptive catalogue; there are more urgent affairs. (2) The task is now approached with a deeper sense of responsibility, for it is recognized that every peculiarity does not mean a new species. Many peculiarities are transient somatic modifications which are reimpressed on successive generations as long as the same incident factors are operative. Other peculiarities are not beyond the range of the variations which occur within the limits of one family, the prog- eny of one pair. It is a pious opinion that a new species should be based on numerous specimens, and that something should be known of the genetic behaviour of its peculiarities. (3) More attention is being given to the occurrence of variations and mutations, which, instead of being put in a corner as incon- veniences, are now regarded as of no less interest than the rank and file of the species. Polymorphic species, like the ruff, are more interesting biologically than relatively fixed species like the pheasant. (4) But along with this increased and more ana- lytic attention to variations, which have to be studied from the point of view of genetics, and if possible experimented with, there is an increased recognition of the other side the dis- continuity or apartness of species. It is difficult to maintain that the category species is univalent throughout, e.g. that it has the same value among micro-lepidoptera as it has among birds, but in all sound taxonomy a species is marked by its definite specificity. The tendency is toward a recognition of fixity as even more characteristic than flux. In many cases it seems safe to say that species differ in the presence or absence of unit characters, or in the arrangements of their unit characters, whose gametic factors exhibit Mendelian behaviour. (5) If species differ, like a series of organic compounds or like a' series of related minerals, in their internal constitution, then it is no longer necessary to insist on the utilitarian value or adaptive significance of the specificities. That specific characters are often adaptive seems undeniable; that they need have this value is highly improbable (see Bateson, 1913).
Taxonomy Criticised. Supposing the most satisfactory de- scription of a new species, based on numerous specimens living in known environment, and supposing even some acquaint- ance with the genetics of the species, we must face the further question whether zoological work at this Linnaean level is worth while. It is in a way hodman work, though few do it very well. It is very quantitative work, adding leaves to inherited mono- graphs. It tends to preoccupy the investigator, leaving him with little time or inclination for more adventurous pursuits. It leads to little in itself. Thus there is a widespread conviction that the day of the pure systematist is past, and should be declared as past.
But this is perhaps not more than a wholesome rebound from an unsatisfactory species-mongering. In any case there is another side to the criticism. If a region is to be surveyed in a business-like way, e.g. the Zoological Survey of India initiated and directed by Dr. Nelson Annandale, there must be a per- fectly precise census. For practical purposes it is indispensable that there be no vagueness about the species of (say) mosquito and tick. The meticulous thoroughness, sometimes scoffed at by the impatient, may save the lives of thousands. Moreover,
