But it is hardly necessary to point out that all this is based on the assumption of =Uniformity= in its most extreme type, a doctrine which not only denies that these living forms are merely the =lucky survivors= of tremendous changes in which their contemporaries perished, but which in essence is taking for granted beforehand the very point which ought to be the chief aim of all geological inquiry, viz., How did the geological changes take place? It would not be considered a very scientific procedure for a coroner, called upon to hold a _post mortem_, to content himself with interesting statistics about the percentage of people who die of old age, fever, and other causes, while there was clear and decisive evidence that the poor fellow had been =shot=. In this case, as in geology, it is not merely the result that is wrong, but the whole method of investigation. For, as in the latter case we don't want to know how people generally die, but how this particular person actually did die, so, in our study of geology, we do not wish to know merely the rate at which changes of surface and extinctions of species are now going on, and then project this measure backward into the past as an infallible guide, but we wish to know for sure just what changes of this nature have taken place. A true induction is, I think, capable of deciding very positively whether or not the tools of nature have always worked at the same rate and with the same force as at present; and this method of arranging the fossils in supposed chronological order on the percentage basis mentioned above, is only an extreme form of methods claiming to be inductive which in this age of the world ought to be considered a shame and a disgrace, because, as Howorth says, they are based, "not upon induction, but upon hypotheses," and have "all the infirmity of the science of the Middle Ages."