All this information is gained slowly, often very slowly, and with much labor and pains. Rarely or never is it the case that all the information obtainable concerning any one kind of an extinct animal is furnished by a single specimen. Skeletons are very seldom, perhaps never, found quite complete, with all their parts in their natural positions; and the nature of the matrix inclosing them usually prevents a study of all parts of any specimen. If a newly discovered fossil is widely different from the corresponding parts of any creature previously known, whether living or extinct, we cannot infer very much from a few bones as to what the remainder of the skeleton is like. Such inferences or guesses in the past have often resulted in grievous error, and self-respecting paleontologists are now very reluctant to speculate much concerning extinct animals from fragments of a skeleton, no matter what those fragments or bones may be; future discoveries are sure to reveal errors. It is, therefore, only by the accumulation of much material, and by the careful study and comparison of all known related animals, that reliable conclusions can be reached. Often it requires scores of specimens to determine the exact structure of a single kind of animal, and, as the collection and preparation of fossil skeletons are tedious and expensive, our knowledge sometimes increases very slowly. In recent years, however, there have been many more students of extinct backboned animals than formerly, and there are now many museums and universities which spend annually large sums of money in the collection and preparation of such fossils. This greater activity of the last twenty years is bringing to light many new and strange forms, as well as completing our knowledge of those previously imperfectly known. It is commonly, but erroneously, believed that the bones of extinct animals are usually found in excavations made for the purpose. It is true that not a few specimens of fossils have been discovered in excavations made for other purposes, such as railway cuttings, quarries, wells, etc., but if no others were found our knowledge of the animals of the past would be very meager indeed. Fossils are, for the most part, found by deliberate search over the denuded rocks in which they occur. Methods of search and collection will best be understood by the following description of the noted fossil-bearing rocks of western Kansas. [Illustration: FIG. 1.—A characteristic chalk exposure in western Kansas, a hundred acres or more in extent.]