172 



KNOWLEDGE. 



[August 1, 1898. 



pebbles ofDen lie well separated in a dark green-grey 

 matrix of coarse sand. Among them are granites, and 

 quartzites, and compact flinty lavas — evidences of the 

 long gap that divides them from the underlying igneous 

 and metamorphic series. The beds below are more highly 

 tilted, and everywhere the unconformity is a marked one. 

 The ancient land, the floor of the country, at whatever 

 period it first arose, was pierced by igneous masses, and 

 was baked and altered ; its shales became slates and 

 schists, its limestones marbles ; and the invading rocks 

 had cooled down in the form of granite before the period 

 of denudation occurred that formed the conglomerates of 

 Killary Harbour. In this area, then, we have the old 

 land of unknown age, perhaps even a remnant of the 

 pre-Cambrian chains ; it comes rapidly into view as we 

 go south along the Joyces' Eiver, and culminates in the 

 quartzite ridges of the Maam Turk Mountains and the 

 Twelve Bens. Then we have the grand scarps and cirques 

 of Formnamore, and the impressive and close-set group of 

 summits that form the Mweelrea range on the north side 

 of the fjord ; these represent the thick deposits on the 

 Ordovician and SOurian shoreline, in a sea that lay open 

 to the north.* Then came the widespread Caledonian 

 movements. In Devonian times, our highland was thus 

 already well established, looking down into lake-basins 

 that lay many miles away on the north-east and the east ; 

 and then a second great subsidence brought it below the 

 Carboniferous sea, and for a time saved it from denudation. 

 We do not know when it made its next appearance ; but 

 probably the covering was worn away from its bolder 

 summits soon after the uplift of the Hercynian chains. 



The complex details have even yet to be worked out ; 

 but the district is clearly one of those that hive remiined 

 highlands by the force, as it were, of pure tradition. Many 

 parts of the west of Scotland, the English Lake District, 

 the volcanic precipices and arrtef: of Wales, have similarly 

 proved their powers of resistance, and tbeir tendency to 

 reappear as knots through any covering forced upon them. 

 In this they resemble Suess's favourite " horsts " of 

 Central Europe ; and it is easy to predict that they will 

 remain as the bare skeleton of Britain, long after the films 

 of the London Clay, the Chalk, or the soft Triassic sand- 

 stones, have been washed away into the North Sea and 

 the Channel. 



The vitality of Ireland similarly lies in the great ribs of 

 the west, holding their own against the Atlantic and its 

 warm soft air. Even if the oscillations of the continental 

 edge again submerge her, Connaught will still exist, a core 

 of mountains, in the depths. 



At present, the stratification of the series of rocks out 

 of which the hills have been carved is still well marked 

 around us. The north scarp of the Formnamore group 

 is seen in Fig. 1, where the terraces of Ordovician and 

 Silurian strata form a feature of the steep hillside. In 

 Tonalee, again, above the Maam valley, the unconformity 

 between the old " Dalradian " series and the overlying 

 Ordovician conglomerates is traceable even at a distance. 



North of Killary Harbour, a road traverses the Mweelrea 

 Mountains by a low pass, and brings us into the most 

 exquisite landscapes of the range. For two miles, along 

 the west shore of Lough Doo, the purple crags of Glan- 

 cuUin, ledged with green, rise some two thousand five 

 hundred feet above the water, and the steep southerly 

 dip of the beds adds to the serration of the mountain-face. 

 The scene is even finer if we turn up the valley td the 

 east, and view it from the moorland level, with Doo 

 Lough answering to its name, and lying black below us. 



• See Sir A. Geikie, Ann. Sep. Oeol. Survei^ for 1896, p. 51 



Siill further to the east, the Dalradian floor is met with 

 in the lonely bogland through which Lough Tawnyard 

 extends. Here the fine cirques and mountain-crests are 

 still formed of the upper stratified series, the masses of 

 which rise in great outliers upon the worn-down edges of 

 the older rocks. 



We have already hinted at the proaiinence of the floor 

 itself, the " pre-Ciledonian " ridges, in the fine region of 

 the Twelve Bens of Connemira. If we follow the fjord of 

 Killary down to its mouth, we can look across thirty miles 

 of blue but restless water, until the eye is caught by the 

 huge cone on Achill Island ; this is cut in half on its west 

 side, where its cliffs drop more than two thousand feet into 

 the sea. Here, again, the older series still asserts itself, 

 bared from any covering of Silurian or Carboniferous 

 deposits. 



As we return eastward up the inlet to Leenane, the 

 strange aspect of the fjord itself is impressed upon us, 

 and our thoughts are transferred to the most recent of 

 geological epochs. The deep groove-like nature of the 

 hollow that is occupied by the sea is well seen as we 

 climb the moors upon the south ; and thence we look, 

 down into the water, where the fresh tide has covered all 

 the sandbanks, and has pressed back the flow of the river 

 with a curving line of foam. This groove is ten miles 

 long, and is rarely more than half a mile in width. At 

 its head (Fig. 1) it passes continuously up into the 

 valley of the Erriff, which is bounded by the same steep 

 green or clifi'-set walls. 



At the hamlet of Aasleagh, where we reach the actual 

 river, there is a pleasing little waterfall over a step in 

 the valley-floor ; and below it there is a second tumbled 

 fall, where the seaweeds and the wild-flowers meet, and 

 where the former clothe all the rocks out in the stream. 

 Here we see the excavation of the valley still going on. 



But this is only a feature of low water. At high tide 

 the sea reaches the bridge of Aasleagh, and the clump 

 of fir trees looks down upon ephemeral waves. All 

 evidence of the activity of the river on its rocky bed has 

 disappeared. 



There is at this point an obvious connexion between the 

 valley and the marine inlet. As the river, in the inter- 

 spaces of low water, erodes its bed, the sea can spread 

 further, though imperceptibly, inland at each successive 

 tide. Will not this cutting baik of the head of the fjord, 

 year by year, account for the long intrusion of the sea 

 upon the land ? 



In this case, however, and still more strikingly in that 

 of other fjords on our coasts, examination of the Admiralty 

 charts will show that stream-erosion alone is not sufficient. 

 Killary Harbour happens to deepen fairly uniformly from 

 Aasleagh to the open water ; but its depth at its mouth is 

 twenty-two fathoms, while off Mweelrea and Salruck it is 

 still ten fathoms. The cutting action of the river cannot 

 be responsible for excavating a groove of this depth, a 

 great part of which lies below the level of low water. 



But if the land were uplifted, the bottom of the inlet 

 would become a portion of the ordinary valley-floor. The 

 river would reach the sea between Inishbofin and Inish- 

 turk, and would be able, above this point, to deepen its 

 valley until a level slightly below low water was attained. 

 The history of Killary Harbour is no doubt embodied in 

 the reverse of this suggested process. The h;nd at no 

 distant time stood distinctly higher above the sea, and the 

 Errifl' River, from Aasleagh down, had a fairly rapid fall. 

 The great groove, in fact, severing Mweelrea from Ben- 

 choona, is an ordinary river-valley, cut by a stream that 

 started in pre-glacial times. The floor of this valley 

 gradually approached the sea-level, the level of no ex- 



