and is seldom to be recommended. There can be no doubt that badly formed and inappropriate names have been a just cause of much of the criticism of systematic work. The correcting of errors is always an ungracious task, especially where the errors are apparently trivial and obviously unintentional, and this is particularly true in systematic zoology. No reasonable person will, however, question the necessity of it or refuse to accept the correction if it is justified. But in such work, accuracy of statement, soundness of judgment, clearness of reasoning and perfect courtesy are required to an unusual degree. These qualities are, of course, essential for the best scientific work of any kind, but they are particularly so in systematic zoology, where so much work is of an elementary nature. The lack of one or more of them has depreciated the value of many an elaborate monograph. On the other hand, their presence does not alone guarantee the worth of a piece of zoological research. And this suggests the second principle I desire to emphasize.
To be of real worth and permanent value, the systematic study of any group of animals must take into account, so far as they are known, the pre- and post-natal development, the geological history and the geographical distribution of the species which compose it.
A systematic study of any group of animals which considers only the adults, even if the morphology, physiology and habits are all taken into account, can not be regarded as complete. Nature would indeed be a puzzle and interrelationships a hopeless snarl, if the stages of development were not discoverable or were meaningless. It is incredible to me that any zoologist, who has examined the evidence at all, can deny the existence of stages in both pre-and post-natal development or question the fact that those stages have some meaning. And I am unable to believe that we can even approximate the true history of any group of animals so long as those stages are ignored. Equally important is the paleontological evidence and to ignore it when it exists in any appreciable amount is indefensible and may result in deplorable error. For some reason the relationship of systematic zoology to geographical distribution has been more generally recognized than its relation to paleontology, but it seems to me obvious that in the diversification of the animal kingdom, the element of time has been fully as important as that of space. In neither case, however, is it justifiable to assume that quantity (of either time or space) is necessarily correlated with specific qualities. Discontinuous distribution, either geologically or geographically, ought never to be regarded by itself as a differential character. Either may be used as an additional character for a group otherwise structurally distinguishable, but no species (or other group) ought ever to be recognized whose identity can not be determined without knowing the locality or the geological horizon from which it came. To act on any other principle can not fail to lead to