374 SAND-FLOOD IN THE HEBRIDES. 



of Harris, and tried extensively upon his estate. It con- 

 sists of planting small bundles of Arundo arenaria, at 

 distances of about a foot and a-half. These take root, 

 and prevent the drifting to a certain degree. But often 

 vegetation is tardy in establishing itself, and if the turf 

 plan be not considerably more expensive, it seems pre- 

 ferable, because it very effectually prevents the drift, 

 and moreover, produces excellent pasture ground ; the 

 former of which indications, the planting system, does 

 not completely effect, and the latter in a very imperfect 

 degree." 



We may add, as this subject is a very interesting 

 one, that further details, in regard to the moving sands 

 of Scotland, will be found, on consulting the Statistical 

 Account of Scotland, vol. xx. p. 220. In the Appendix 

 to the Account of the parish of Dyke, vol. xx. p. 228. 

 et seq. there is an account of the Sand-Hills of Maviston, 

 which overwhelmed the barony of Coubine, as mentioned 

 in Mr Ritchie's communication. In vol. xix. p. 622. 

 is a notice of the shifting of two hills of the Ma- 

 vieston Range 500 yards in twenty years. In vol. xxi. 

 p. 207-5 is a notice of some hundred acres in Duffus 1 pa- 

 rish covered uiree feet deep by drift sand ; fourteen inches 

 accumulating in one night. In NeilTs Tour in Orkney 

 and Shetland 1804, it is observed, that, in the neighbour- 

 hood of the Castle of Noltland,in Westra, much havoc has 

 been done by the blowing of the sand. No measures are 

 there employed for putting a stop to this kind of devas- 

 tation. In the 6th volume of the Highland Society's 

 Transactions will be found a report of the operations car- 

 ried on in Harris, and alluded to in Mr Macgillivray's 

 communication. And in Dr Walker's Account of the 



