In most books upon ornithology--in this one, for instance--the scheme of classification is set forth in the shape of a list beginning with one particular group and ending with another. This is merely due to the physical properties of sheets of paper. A linear scheme is really an impossibility; to represent classification properly we want a solid diagram, showing how from a root-stock branches arose and pushed their way in every direction. Another defect of the linear scheme is that we must begin somewhere and end somewhere. In this book we begin with the Passeres and end with the Parrots; others start with the Accipitres, in spite of the protest of Michelet against placing the cowardly, flat-headed, stupid hawks at the summit of bird creation. It doesn’t matter where we begin or where we end as long as we carefully bear in mind that a linear classification is only a convenient way of briefly stating certain facts, and that it does not pretend to be a copy of Nature. An alternative method of expressing the facts of structure in space of two dimensions is the Stammbaum, originally made in Germany; but this inevitable tree of life is open to the serious objection of undue dogmatism; and besides, it must be inaccurate, as it is not in three dimensions. A given naturalist may have strong reasons for believing, let us say, that the Struthious birds represent the lowest bird stock, from which arose in a regular series of branches, independently, and alternately from one side or the other, the various groups into which we divide the class in the present book; if so, then the Stammbaum is easily constructed. But the general consensus of opinion is that the inter-relationships of the different groups cannot be expressed with so much simplicity. It is clear that, in any case, the most modified offshoots must occupy the highest branches of the tree, and that we may in a linear scheme conveniently begin or end with them. But it is impossible to arbitrate as to which group is the most specialised. It is, on the whole, agreed that the Ostrich tribe have retained more primitive characters than other birds; but is the elaborate voice-mechanism of the Nightingale, or the almost human intelligence of the Raven or Parrot, to rank first as evidence of high position, _i.e._ specialisation, remoteness from the original stock? This is a matter about which everybody can legitimately have an opinion; and we cannot at present formulate a creed--for those, that is to say, who are acquainted with the facts.