104 



THE FORMS OF WATER 



quickly than the western side. 



24. SUGGESTION OP A NEW LIKENESS OF 

 ^LACIER MOTION TO RIYEU MOTION. 

 CONJECTURE TESTED. 



173. Here we have cause for reflection, and 

 facts arc comparatively worthless it' they do 

 not provoke this exercise of the mind. It is 

 because facts of nature are not isolated hut 

 connected, th:it science, to follow tiiem, must 

 also form a connected whole. The mind of 

 the natural philosopher must, as it were, ba 

 ii web of thought corresponding in all its 

 fibres with the web of fact in nature. 



174. Let us, then ascend to a point whi^li 

 commands a good view of this portion of ihe 

 Mer de Glace. The ice-river we see is not 

 straight hut, curved, and its curvature Isfrotn 

 the Mootauvert ; that is to say, its convex 

 side is east, and its concave side is west (look 

 to the sketch). You have already pondc;ed 

 the fact that a glacier, in gome respect*, moves 

 like a river. How would a river move 

 through a curved channel ? This is known. 

 Were'tha ice of the Mer de Glace displaced 

 by water, the point of swiftest motion at the 

 Montanvert would not be the centre, but a 

 point east of the centre. Can it be then that 

 this "water rock," as ico is sometimes 

 called, acts in this respect also like water ? 



175. This is a thought suggested on the 

 spot ; it may or it may not be true, but the 

 jjieans <>f testing it are at hand. Looking up 

 the glacier, we see that at ks Fonts it also 

 bends, but that there its convex curvature Is 

 toward the western side of the valley (look 

 again to the sketch). If our surmise be tiue, 

 the point of swiftest motion opposite les 

 Fonts ought to lie west of the axis of the 

 glacier. 



176. Lt j t us test this conjecture. On July 

 25th we fix in a line across this portion of the 

 glacier seventeen stakes ; every one of them 

 has remained firm, and on the 2(jth we lind 

 the motion for 24 hours to be follows : 



FOURTH LINE : D D' UPON TIIE SK.ETCU. 



East West 



Htako 1 2 3 456789 10 11 12 13 14 15 



Inches....? 8 13 15. 16 19 20 ^1 fcl 23 23 sil 22 IT 15 



177. Inspected by the naked eye alone, the 

 stakes 10 and 11, where the clacier reaches 

 its greatest motion, are seen to be considera- 

 bly to the west of the axis of the glacier. 

 Thus far we have a perfect verification of 

 Wie guetss which prompted us to make these 

 measurements. You will here observe that 

 the " guesses" of science arc not the work 

 of chance, but of thoughtful pondering over 

 antecedent farts. The guess is the " induc- 

 tion" from the facts, to be ratified or ex- 

 ploded by the test of subsequent experiment. 



178. And though even now we have ex- 

 ceedingly strong reason for holding that the 

 point of maximum velocity obeys the law of 

 liquid motion, the strength of our conclusion 

 will be doubled if we can show that the 

 point shifts back to the eastern side of the 

 axis at another place of flexure. Fortunate- 

 UT such a place exists opposite Trelapqjte. 



Here the convex curvature of tne vallsy 

 turns again to the east. Across this portioa 

 of the glacier a line was set out on -July 28th, 

 and from measurements on the Ulst, the ratu 

 of motion per 24 hours was determined. 

 FIFTH LIN T E : E E' UPON THIS SKKTCII.-] 

 West East 



Stake 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 



Inches.. .11 14 13 15 15 16 17 19 20 19 20 18 16 15 13 



179. Here, again, the mere estimate of 

 distances by the oye would show us that 

 the three ^ takes which moved fastest, viz. 

 the 9th, 10th, and lllli, were all to the east 

 of thu middle line of the glacier. The dem- 

 onstration that the point of swiftest motion 

 wanders to and fro across the axis, as the 

 flexure of the valley changes, is, therefore 

 shall I say complete V 



180. Not yet. For if surer menns are open 

 to us we must not rest content with estimates 

 by the eye. We have with us a surveying 

 chain : let us shake it out and measure these 

 lines, noting the distance of e^ery stake 

 from the side of the glacier. This is no easy 

 work among the crevasses, but 1 confide it 

 confidently to Mr. Hirst and you. \Ve can 

 afterward compare a number of .slakes on the 

 eastern side with the same number of stakes 

 taken at the same distances !r-m the western 

 side. For example, a pair of stakes, one ten 

 yards from the eastern side and the other ten 

 yards fiom the western side; another pair, 

 one fifty yards from the eastern side and the 

 other fifty yaids from the western side, and 

 so on, can v be compared together. For the 

 sake of easy reference, let us call the points 

 thus compared in pairs, eqniraient points. 



181. Tin re were five pairs of such points 

 upon our fourth line, D 1)', and here are 

 their velocities : 



Eastern points ; motion in inches.. 13 15 16 18 SO 

 \Vertern points " " -.15 17 1'2 23 23 



In every rase here the stake at the western 

 side moved moie rapidly than tlu equivalent 

 slaKO at the eastern side. 



182 Applying the same analysis to our 

 fifth line, E E', we have the following series 

 of velocities of three pairs of equivalent 

 points : 



Eastern points ; motion in inches 15 18 19 



Western points " 13 15 17 



183. Hern the three points on the eastern 

 side move more rapidly than the equivalent 

 points on the western side. 



184. It is thus pioved : 



1. That opposite the Montanvert the east- 

 ern half of the Mer de Ulace moves more 

 rapidly than the western half. 



2. That opposite lex Pouts the western half 

 of the glacier moves more rapidly than the 

 eastern half. 



3. That opposite Ti elaporte t .e eastern half 

 of the glacier again moves more rapidly than 

 the western half. * 



4. That these changes in the place of great- 

 est motion are determined by the flexures of 

 the valley through which the Mer de Glace 

 moves. 



