1876.] ^OJ [Frazer. 



On the Ain (one of the affluents of the Upper Rhone) it is over 120 

 centimetres. 



Up ihe Seine from Paris about 28 p. c. of all the water which falls in what- 

 ever form tinds its way into that river (within that hydrographic basin); 

 53 p. c. in the basin of the Saone near Trevoux. For the Garonne above 

 Marmande and for the Auxois which flows over impermeable marls it 

 rises to 65 p. c. The percentage increases in schistose and granitic regions. 



It is noticeable that the rivers which flow over calcareous beds do not 

 exhibit a large amount of limestone in their debris, as for example the 

 Loire. This is attributed to the softness of calcareous fragments and their 

 inability to resist triturition. [It may be remarked that this explanation 

 will only apply to the deposits at considerable distances from the edge of 

 the calcareous formation, for otherwise the waste would be constantly re- 

 newed.] 



400 metres from the mouth of the Lez the sandy deposits contain over 

 60 p. c. of carbonate of lime. At the mouth it consists of 23 p. c. carbonate 

 of lime, with great numbers of flakes of green mica. 



The HeraultGOO m. from its mouth deposits a reddish gray gravelly sand, 

 consisting of hyaline quartz with orthoclase, feldspars and granitic debris, 

 much argillaceous schist, reddish brown mud, gray mica schist, greenish 

 quartz, sand, and some tablels of silvery mica. 



Magnetic sand is found here in large quantity and is derived from the 

 igneous rocks over which the river flows. 



These few notes will serve to convey an idea of how this part of the sub- 

 ject is treated. 



By far the most striking, as the most valuable, part of M. Delesse's 

 work is to be found in the elaborate charts which accompany it. The 

 character of these charts as .specimens of map printing is very high, 

 although it is questionable if the condensation of a great number of un- 

 allied statistical facts upon one sheet is favorable to the easy comprehen- 

 sion of such charts. A better plan would be to reproduce the same out- 

 line several times with different and strongly marked coloration or shading 

 than to construct all curves of every kind upon one frame work, thus causing 

 them to cross and recross one another to the bewilderment of the eye. 

 Another dilBculty for the reader arises out of the method adopted by M. De- 

 lesse for conscientiously representing all his facts upon a single sheet. Had 

 the several kinds of curves been strongly portrayed they would so attract 

 and confuse the eye as to entirely destroy the effect of the map as a whole. 

 To avoid this the lines have been made so exceedingly fine, that if any one 

 try to follow one of these curves he will need either an exceptionally 

 good ej'e or a magnifying glass; in either case the character of a boundary 

 between a territory on one side and a territory on another fails; for the mo- 

 ment the observer seeks to take in the entire curve at a glance it becomes 

 invisible. 



The datum plane is sea level. Heights are registered in the metre curves 

 and depths by similar means. But in approaching the mountainous regions 

 of France, one half the contour lines are omitted, lest they should shade 



