l88 EFFECTS OF FORESTS ON SPRINGS AND RIVERS. 



" If we omit from this table the observations made with the pegel 

 on the Ehine at Germersheim, — as the lowering of the water level 

 there, together with the deepening of the river bed, is mainly attri- 

 butable to the extensive rectifications of the river by cuttings, and only 

 in a comparatively small degree to a diminution of the delivery, — we 

 find that the greatest lowering of the water level is on the Danube at 

 Orsova, amounting to from 46 to 55 inches ; on the Vistula at Marien- 

 werder from 26 to 43'9 inches ; and on the Rhine at Emmerich from 

 20 to 24:'8 inches ; and on the other pegels of the rivers named from 

 6 to 18 inches. 



" In the Elbe the falls of the lowest and of the mean water levels 

 are 15'76 and 16'28 ; and they might indeed have been considerably 

 greater had it not been that since 1842, in consequence of the great 

 silting up of the river bed with sand, the lower water levels have been 

 elevated from 13 to 22 inches. 



" The lowering of the water level becomes so much the greater the 

 larger the river is, and the nearer to the mouth of the river in the sea 

 that the pegel is situated, and this for the reason that the region 

 drained by the river is the greater, and that it is the sum of the reduc- 

 tions in the levels of all the brooks and streams and lesser rivers 

 which have flowed into the nmin stream which is there represented 

 in this greater lowering of the river level. 



" In regard to the high floods of the said rivers, we see from the 

 diagrams submitted that, with the single exception of the station at 

 Orsova, in all of them floods have latterly occurred more frequently 

 and reached a greater height than in the earlier years in which the 

 observations were made, from which it appears clear as day that in 

 these rivers at present, at times of high flood, a much greater 

 quantity of water is delivered by them than in the earlier years. 



" From the diagrams representative of the high floods, it is further 

 apparent that in the earlier times of observation the rise of water, in 

 consecutive years, was more uniform ; whereas, in the later decades 

 of the periods embraced by the observations, in one year there 

 occurred a very great flood, and in the year before or the year 

 after only a very insignificant rise of water ; from which it follows 

 that the alternation of very dry with very wet years occurs more 

 frequently and to a much more marked degree now than for- 

 merly. 



"This is particularly apparent in the tabulated observations on 

 the Elbe and the Vistula. 



" The cause of this remarkable phenomenon lies evidently in this, 



