GEOGRAPHY OF CENTRAL FLORIDA 227

North America. (If only nesting birds are counted we have about one-fourth.) The exact number is and always will be indeterminate, on account of differences of opinion as to what constitutes a species, if nothing else.t Another reason is that there are quite a'number of birds that feed in the ocean near by but rarely if ever nest on our coasts, or which pass over in their annual migrations between North and South America and seldom stop, and some whose usual migration routes lie considerably to the eastward or westward but are occasionally blown out of their course by storms and forced to land.' Then too there must be many which barely reach our limits from the north or south, and whose ranges are not yet known with sufficient exactness to indicate whether they occur within the arbitrary limits of this work or not. But all these uncertainties should not materially affect the statistical observations which follow.
 The birds of North America are divided into two great groups: water birds, comprising (according to Chapman) 9 orders and 29 families, and land birds, with 8 orders and 37 families. The former are the more ancient and primitive types, and seem to be most characteristic of regions that are geologically young, while the latter have evolved so recently that there are comparatively few fossil records of them, and they are most abundant in regions that have been dry land for ages. About 46% of the birds (species,
not individuals) in central Florida are water birds, as compared with 42% in eastern North America, 38% in the whole United
States and Canada, and only io or 12% in the whole world.* But
the water birds as a rule have wider ranges or migrate more than the land birds, so that even if other things were equal they should
 $As there are more bird students than species of birds in civilized countries, the temptation is .strong to keep drawing finer distinctions, making slight differences the-basis of subspecies, and elevating subspecies to the rank of species from time to time. Birds of widely distributed species that do 'not migrate much are apt to be a little smaller and darker in Florida than farther north, and already quite a number have been separated for that reason, and doubtless moire will be hereafter. Some of the mammals show the same sort of variation, as was pointed out by Dr. Allen in the paper previously cited.
 *This high percentage of water birds in new lands seems analogous to the high percentage of monocotyledons among flowering plants in the same areas. See a -statistical method for comparing the age of different floras, in Torreya for December, 1905. Also 3d Ani. Rep. Fla. Geol. Surv., p. 357.