112 LITERATURE ON TORRElN'ia. 



taken place, and in whicli the most eminent savants have taken part, have 

 been confined to that of causes which could act on the delivery ; and the 

 whole discussion has come back to that of the permeability or imperineability 

 of the soil. Even the effect of forests has not been studied beyond what 

 the consideration of them from this point of view required. All of the 

 researches which have had for their aim to enable us to combat inundations, 

 have had no other object but an action to bear on this delivery. All who 

 have written on the subject have reasoned and made their calculations as 

 if, at the time of an excessive flood, nothing was occurring but an augment- 

 ation of the volume of the current, without any variation in the hydraulic 

 law by which it was regulated. I have no hesitation in saying I con- 

 sider this way of looking at the subject erroneous ; and it is at this 

 point that I take my departure from those who have preceded me — on a 

 new enquiry. 



" From my point of view there is seen to be something more than simply 

 a variation in the delivery. At the time of a great flood, when a current-- 

 be it great or small — bears along considerable solid masses, consisting of 

 earth and stones of all sorts and sizes, a peculiar phenomena of special 

 importance is evolved. This is a perturbation, more or less marked, in the 

 progress of the current, and in the laws by which it is regulated ; and this 

 it is which I call the torrential phenomeiwn, or, if a word must be created 

 under which to speak of it, the torrentiality — an action of perturbation 

 which is the greater in proportion as the secondary causes by which 

 it is produced — namely, the solid matters borne along — are the more 

 considerable. 



" From this point of view, the most furious torrents of the Alps are seen 

 to be only extreme cases of a general phenomenon, which is produced more 

 or less imperceptibly, or more or less distinctly marked, in all currents of 

 water which are not perfectly tranquil in their flow. 



" The characteristic effect of this perturbation is an instability in the 

 course of the stream. 



" When a current of water does not bear along solid matter, whatever may 

 be the volume of water, the flow is effected with great stability in accord- 

 ance with hydraulic laws. Sudden variation in the delivery, in raising or 

 lowering the level, produce variations in the rapidity of the flow ; from this 

 there is thus a certain consequent perturbation ; but the action of gravita- 

 tion, in its omnipotence, being constant, and this accommodating itself to 

 the resistances due only to friction, the stability of the stream tends 

 uninterruptedly to maintain itself. To a rise of level there being a corres- 

 ponding increase of rapidity of flow, it is rarely the case that such waters 

 rise higher than the banks. 



" The perturbation produced by solid material borne along is, on the 

 other hand, very serious. If the substance of the cuiTent be very greatly' 

 changed in conBistence — for example, if for a limpid water, possessing all 

 its fluidity, there be substituted a viscous liquid — if, further, the torrent bo' 

 required to perform the mechanical work of conveying a certain quantity of 

 solid matters — the conditions are greatly modified. In the first place we 

 have no longer simply pure water, but water subject to every degree of 

 variation in so far as fluidity is concerned ; and thus the work of transport 

 imposed on the current developes resistances which are subject to every 

 degree of variation. From this birth is given to an extreme instability in 

 thie current, or in other phrase, to torrentiality. 



