Origin of " Dirt-bands " in Glaciers. 135 



those laws in obedience to which the ribbon-structure origi- 

 nates. For this reason, the curve of the bands is, like that 

 of the structure, found to be elongated low down the glacier, 

 and compressed as we approach its source. We have thus 

 only to account for the existence of bands of different kinds 

 of ice, the form of curve of those bands being explained in the 

 same way as that of the ribbon- structure. 



It may be observed, that the superior distinctness of the 

 dirt-bands, as we proceed lower down the glacier, is not ne- 

 cessarily an evidence that the bands of porous and compact 

 ice are there more decidedly developed, but only that they 

 are more distinctly apparent. And this, I imagine, arises 

 from the fact, that the lower ice has been washed over for 

 years ; and thus the pores have become more discoloured by 

 the deposit of drift than the pores of the corresponding 

 porous ice above. 



It was the remarkable similarity of the alternating bands 

 on the mud-slide already described, to the " dirt-bands" on 

 Professor Forbes' map of the Mer de Glace, that induced me 

 to take an interest in the matter, and make a drawing of the 

 phenomenon. In the first instance, the curved bands were a 

 mystery to me ; and I could not venture to found any argu- 

 ment on a mere analogy of appearance. The second mud- 

 slide; however, seemed to shew me another step in the pro- 

 cess ; and, having explained the one from the other, I was 

 led to ask myself whether the phenomena, to which one class 

 of viscous fluids appeared to be subject, might not be com- 

 mon to another ; — in other words, to a glacier. 



In our first mud-slide we observe, firsts the occurrence of 

 curved bands ; and, secondly^ a difference of consistency in 

 those bands. Our second mud-slide shews the origin of those 

 curved bands, as far as mud-slides are concerned, to be the 

 previous existence of ridge or wrinkles. Turning, on the 

 other hand, to the glacier, we find curved bands of difi*erent 

 consistency and similar appearance, which I have called re- 

 latively bands of porous and compact ice. 



We may then fairly ask, — If, in one species of viscous fluid, 

 alternate bands are derived from pre-existing ridges, why 

 should not analogous bands in another species of viscous fluid 



