340 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



a relatively high stage of civilization. In 1883 a two-story log house 

 surrounded by an inclosure was found in Drumkelin Bog, County 

 Donegal ; it was 12 feet square and 9 feet in height, and a roadway 

 led to it across the bog. Both house and roadway were entirely con- 

 structed of oak, no fir being used. With the hut were found a stone 

 chisel and a flint arrowhead. Beneath the floor were 14 feet of 

 bog, and above the floor 26 feet. Other roadways or " cashes " have 

 been found, e. g., at Ballyalbanagh, County Antrim, above 4 feet of 

 black turf and below 5 feet of flow bog; at the base of the black turf 

 were oak logs and stools, and on the level of the road fir stools. None 

 of the latter were beneath the road, which was built entirely of oak. 

 A neolithic ax was found near the road. These discoveries of an ad- 

 vanced neolithic civilization on the forest layer confirm our inference 

 that it is younger than the 25-foot raised beach, for in and on the 

 latter are found numerous neolithic implements of a far more primi- 

 tive type. Probably the peat under the forest layer representing 

 moister conditions corresponds to the 25-foot submergence. 



Many fine relics of the bronze and early iron ages have been found 

 at various depths in the peat; a number of lunulae or collars of 

 gold, bronze swords and daggers and a long rapier of bronze. In 

 a bog at Cromagh, Annoy, County Antrim, was found a woolen 

 garment, with a bronze pin, celt, gouge, razor with leather case, and 

 an ornament of horsehair. Bodies dressed in old-fashioned woolen 

 garments have been found from time to time. 



In interpreting the meaning of this well-marked peat bog suc- 

 cession, we must remember the peculiar conditions of peat formation. 

 The basal layer of oak stools, rooted in the solid rock or even in la- 

 custrine marl, is merely what we should expect, for on any ordinarily 

 well -drained ground the water flows off too rapidly for the growth 

 of peat. The ground is therefore first occupied by a forest; if the 

 region is a lough shore, the trees grow to the water's edge, and as 

 the cutting down of the outlet lowers the level of the water, trees 

 grow at lower and lower levels, encroaching on the old lake deposits. 



The next stage is the formation of leaf mold and perhaps the pros- 

 tration of a few trees; this checks the drainage, and peat mosses 

 grow up between the trees, finally choiring them and entirely occupy- 

 ing the site. In the case of a lake basin the same result might be 

 arrived at by marsh plants choking the waters and finally rising above 

 their level, and so killing the neighboring trees, but this must be 

 more local. So far there is no reason to assume a change of climate ; 

 the desiccation of the peat sufficiently for its use as a dwelling place 

 for man, and for the growth of trees is another matter. The very 

 widespread nature of the phenomenon is proof that the change was 

 not due to artificial draining, and the same fact, as well as the very 



