The Scottish Naturalist. 



161 



dications of where two of these passages had joined. Whether 

 the family lived in the small pear-shaped circle, and used the 

 larger one as an outer yard, or defence-work, can only be a matter 

 for conjecture ; but enough remains to show that both were very 

 strongly constructed. The walls had been built of large glacial 

 stones, probably packed with turf, and may have been from 6 to 

 8 feet thick. Judging from their present appearance, they might 

 have been, at least, the height of a man. The floors, which 

 appear to have been a little lower than the general surface, are 

 paved with stones similar to those of which the walls are con- 

 structed ; but selected more with a view to obtaining as smooth a 

 surface as possible. In regard to the kind of roofing of these 

 " cralls," we can at best form conjectures only. If the smaller or 

 pear-shaped circle was the only part covered, it would have been 

 a simple enough matter ; and even if the larger ones were roofed, 

 it would not have been difficult to find trees long enough to span 

 60 or 70 feet, supported with intermediate poles. The undertak- 

 ings which we have previously noticed would require a greater 

 amount of skill and ingenuity than even this. 



We have, along Donside, a good few of these ancient Erd- 

 houses, roofed across with stones of immense size : and we are 

 not to suppose that the early inhabitants of this country did not 

 possess a natural intelligence which enabled them to turn to 

 account the most convenient materials in the localities in which 

 they lived for constructing the means of protection and defence. 



As many as 200 of these circular foundations can be pointed 

 out along the northern slope of Kinnord ; and we are o 1 uite safe 

 to conclude that these are only stray portions of what was 

 formerly a great centre or metropolis of ancient population. All 

 round the shores of Loch Kinnord, the slopes of Culblean, and 

 the neighbourhood of the Lake of Knockice, traces of ancient 

 workings are to be seen. The hand of the modern improver 

 does not scruple to overturn these old dwellings, transforming 

 their sites into cultivated land, and their stones into modern 

 houses and dykes. This has been done to a very large extent ; 

 but in the case of these more perfect ruins, the stones are not of 

 such a workable kind as the modern mason would care to use ; 

 hence their better preservation. 



Although a very large number of these Pictish dwellings have 

 been obliterated by modern improvements, their lines of com- 



L 



