XXXI. On the theory of the motion of glaciers

Abstract

Almost all the numerous discussions which have taken place during the last twenty years respecting our theories of glacial motion have had for their object the assertion of some particular view, rather than the establishment of a complete and sufficient theory founded on well-defined hypotheses and unequivocal definitions, together with a careful comparison of the results of accurate theoretical investigation with those of direct observation. Each of these views has been regarded, in my own opinion improperly, as a Theory of Glacial Motion . The Expansion Theory ignored the Sliding Theory, though they were capable of being combined; the latter theory was equally ignored by the Viscous Theory, in which, moreover, instead of the definitions of terms being clear and determinate, no definition of viscosity was ever given, though that term designated the fundamental property on which the views advocated by this theory depended. Again, the Regelation Theory is not properly a theory of the motion of glaciers, but a beautiful demonstration of a property of ice, entirely new to us, on which certain peculiarities of the motions of glaciers depend. When we shall have obtained a Theory of the Motion of Glaciers which shall command the general assent of philosophers, no qualifying epithet will be required for the word theory ; it would indeed be inappropriate, as seeming to indicate the continued recognition of some rival theory. If, for instance, it should be hereafter admitted that the sliding of a glacier over its bed and the property of regelation in ice are equally necessary, and, when combined, perfectly sufficient to account for the phenomena of glacial motion, there would be a manifest impropriety, not to say injustice, in selecting either of the terms sliding or regelation by which to designate this combined theory. I make these remarks because I believe that the preservation of the partial epithets above mentioned has a tendency to prevent our regarding the whole subject in that more general and collective aspect under which it is one of the principal objects of this paper to present it. This object must necessarily give to the present paper something of the character of a résumé of what has hitherto been done, whether it be our purpose to adopt or reject the conclusions of others. There are periods in the history of almost every science when its sound and healthy progress may almost as much demand the refutation of that which is erroneous as the establishment of that which is true. I shall not, however, enter into any review of the past labours of glacialists with respect to exploded theories, but shall only notice those more recent researches and speculations which appear to me either to demand refutation as erroneous, or admission into any well-founded theory as correct. This treatment of the subject must necessarily lead to a certain repetition as to results already established, and be also of too critical a character, perhaps, with reference to other results in which I may have no confidence. I can only say that, in the present state of glacial theory, such defects must be inherent in any attempt to present it under a more complete and systematic form than it has hitherto, I think, assumed. It is this circumstance, too, which I would especially offer as an apology for the repetition of certain results which I obtained many years ago. Most of them are abstract mathematical results. They are here obtained by a more general analysis of the problem than that formerly employed, and are introduced as essential steps in the development now offered of the theory of the motion of glaciers.