362 Rev. Edicin Hill — Lotc-ticle Causeivaps. 



springs, "but generally by atmospheric agencies ; by wind, by rain, 

 and by frost, the mightiest of all. But all these lose all their power 

 when once the surface is brought down below water-level, for then, 

 though springs may still issue, they no longer flow down channels ; 

 rain and wind then spend their force on the water ; frost must be 

 fearful if it makes a foot in thickness of ice. Then, indeed, a new 

 set of erosive actions may come into play — waves, tides, currents. 

 In lakes, however, waves are feeble, tides and currents generally 

 absent ; so in such cases the subaqueous bar will last indefinitely. 

 Even on the sea the like may happen in sheltered places, such 

 as fiords, gulfs, or deep bays. In exposed situations where waves 

 are violent, and tides or currents strong, the process of destruction 

 will continue, as it is continuing in Sark. Yet sometimes even in 

 exposed situations, destruction may be delayed or arrested. On 

 shores where much material is being carried along the coast, that is, 

 in places where a groyne would arrest sand or shingle, there such 

 a bar may also arrest them, and instead of being destroyed, may 

 itself accordingly actually grow. A well-known example is the 

 Chesil Bank. (An interesting and valuable paper by Mr. Wheeler, 

 reviewed in Geol. Mag., January, 1897, p. 44, should be consulted 

 in reference to tbis motion of shingle.) Accordingly there is 

 a natural tendency in many situations for bars or causeways to 

 form, at or near the water-level, connecting islets or rocks with the 

 mainland. 



These considerations have some bearing on the force of two 

 arguments adduced in two controversies of old standing. When 

 discussing the antiquity of the present sea-level along the English 

 Coast, it has been argued that the causeway to St. Michael's Mount 

 and the low-tide access to Holy Island have remained unchanged 

 since our earliest records. But the considerations of this paper 

 show that upheaval or depression might be going on and yet these 

 connections suffer no visible alteration through many centuries. 

 Again, in the discussions on the Parallel Roads of Glen Roy, etc., 

 their coincidence with various cols forms a strong argument in 

 favour of the Ice-Dam-Lake theory, and against the theories of 

 submergences. According to that theory, a glen, dammed at 

 its mouth by ice, would fill with water to the level of a col, 

 but at that level the rising waters would find an exit over the col ; 

 the col level would therefore fix the shore-level, so that a road 

 should agree with a col. But on the opposing theory of sub- 

 mergences, why should the resulting shore-lines chance so regularly 

 to coincide with the cols ? This argument is very strong. Yet the 

 considei'ations of this paper show that there would be some tendency 

 for a col to be cut down to the sea-level and no further ; to become 

 as one of these bars ; and so to coincide with a shore-line. These 

 considerations, therefore, to some extent weaken that powerful argu- 

 ment. It is remarkable that Nicol and J. F. Campbell, who both 

 discuss the coincidence of the beaches with the cols, seem neither 

 to have noticed this point. 



