COLOR AND HAUNT AND HABIT. 41 Pacific coast races are darker or more lieavily streaked or barred than anj of their congeners. It is of impor- tance to observe that these differences are shown by joimg birds in fresh phimaoe — evidence that the charac- ters acquired through climate have been inherited. There are many similar eases, but some species seem more easily affected than others, and throughout their ranges are markedly affected by the conditions under which they live. Thus we have nine races of Screech Owl, eleven of Horned Lark, six of Junco, etc. These races, or subspecies, are species in process of formation. The extremes are still connected by inter- mediate or natural links, but if, through any cause, these intermediates should disappear, the extremes would then be left as distinct species. Color and Haunt and Ilaltit. — The relation of a bird's color to its haunts and habits is a complex sub- ject. Any attemj)t at its explanation should be based on so exact a knowledge of the facts in the case, that I can not too strongly emphasize here the necessity for observations in the field. Only a close study of the living bird will justify us in advancing theories to account for its coloration. Many explanations have been offered to account for certain colors and markings of birds, but often, I fear, Avithout adequate knowledge of the bird's habits. I shall speak of only four classes of colors ; they are protective, deceptive, recognition, and sexual colors. Protective colors render a bird inconspicuous in order that it may escape its enemies. Deceptive colors render it inconspicuous in order that it may more easily approach its prey. In both cases the bird should harmonize in color with its immediate surroundings, A survey of the birds of the world shows that on the whole this is true. Thus almost all ground-inhabiting birds, such as Snipe, Plover, Quail, Grouse, Sparrows, 8
