348 OONOIiUSION. 



peoted, that, as stateid, " Between the sources of the Loire and the H6rault 

 the Cevennes are 3,700 feet high. All this surface is composed of granite, 

 impermeable to the rain, and to plant such either with herbage or with 

 trees, is impossible ; " but the answer, thatthe work is being done as fast as 

 money and men and material can be found, and that already, previous to 

 this inundation, all that could be done up to that time had been accom- 

 plished. It is often easy to tell, after an event has occurred, how it might 

 have been prevented ; and it may be that had these inundations been fore- 

 seen, operations which would have to some extent modified or prevented 

 them would have been prosecuted with the vigour called forth by a race 

 against time, in preference to some others which have not been inefiective, 

 but the execution of which might without series consequences have been 

 postponed ; or, at all hazards, grants on a scale of magnitude equalling or 

 exceeding those made previous to the war would have been made, and the 

 difference between these and the amounts actually granted spent exclusively 

 on the valleys and basins of reception drained by the upper waters and 

 affluents of the rivers by which such devastation has been wrought. The 

 legitimate use now to be made of such reasonings is, to prepare for the 

 future in accordance with the suggestions suggested by the past. And this, 

 I have no doubt, will be done. 



The flood of 1875 has proved the most destructive and the most sudden 

 flood of the century; but though floods of such a magnitude are unfrequent in 

 the valley of the Garonne, scarcely do twentyyears pass without the occurrence 

 of a flood of serious importance; and during this century so frequently and 

 regularly have they occurred at such an interval as to suggest as probable 

 the existence of some unknown meteorological law. To some it appears 

 that the periodicity involves a cycle of ten years rather than twenty. 



The Journal des DSbats, writing of the late inundation, says, — " We lately 

 spoke of a probable law regulating the recurrence of rains and floods. 

 What matter whether it be quite correct or no ? If experience prove it 

 true in the majority of cases, why should we not take it into consideration ? 

 All the engineers of the South know that very dry years correspond with 

 sudden deluges of rain, and consequent floods. The unusual dryness of 

 this year ought to have given warning. And the cycle of ten or twenty 

 years now come round should have set people on their guard. Such 

 empirical observations are too lightly disregarded. No hint should be 

 neglected when events of this gravity are at issue. There are memorable 

 dates of which the recollection should never be suffered to die. The cycle 

 appears to differ somewhat according to the region. Thus, for the Loire 

 and the Rh6ne, it seems to be a year later — '46, '56, '66, '76. Are the 

 banks of the Rhdne not in danger next year ?" 



Much interest attaches to the meteorological question involved. Here we 

 have to do mainly with the practical question raised, and the justification 

 of a considerable expenditure of money, of labour, and of thought, on 

 . averting the disastrous consequences following a phenomenon which appears 

 to be, not the result of accidental coincidences, but of frequently recurring 

 coincidences, tha recurrence of which is apparently not only regulated but 

 insured with all tJie certainty attachinf>' to phenomena occui-ring in con- 

 nection with the operation of physical laws. 



Designs of certain measures to prevent such occurrences were proposed 

 after the last great flood in 1855, but they were pronounced too costly ; and 



