eastern slopes, on one of which the white town of Algiers glistens in the sun, descend to the shore at somewhat abrupt angles. Towards the west the declivities are more gradual, and terminate in broad strips of table-land gently dipping for several miles and stretching to the sea. On this side the soil is everywhere covered with a low jungle of brushwood, dwarf oak, palmetto, oleander, lentiscus, cistus, and myrtle, all (when in mass) of dark heathery brown, and tinted also like heather with brighter specks. And so, when standing beside the cromlechs on one of these plateaux, the cottages scattered around, a village in the distance nestling in a hilly nook round a modest church spire, the sea shining in front, the hills behind grouping together, with the bolder peaks of the Atlas towards the west, the landscape presents to us many of the broad features of form, and especially the pervading uniformity of colour which characterize our own highland scenes where the early primeval vestiges have lingered until now.
But with regard to the cromlechs[1] themselves there is certainly (so far as I know) no such extensive group in Great Britain; and I do not remember that, even in the land of megaliths, Britany, so many are now to be found together at any one spot. A few years ago, before some were demolished by the neighbouring colonists of the hamlet of Guyotville, they are said to have been one hundred in number; and at present they may approximately be estimated at fully more than eighty, absolute precision in the enumeration not being attainable from the naturally ruinous condition of some, the recent overthrow of others, and their partial concealment by tangled brushwood. They are spread over an irregular area of probably ten or twelve acres, but they do not stand in equally close proximity to each other over the whole space. In surveying them all from any general point of view they do not suggest the idea of symmetrical arrangement; still, at the north-eastern extremity, towards the outskirts of the group, they take somewhat the form of four nearly equidistant straight rows; and it seems not improbable, although from the numerous gaps and the obliterating vegetation we cannot detect it, that some general plan was followed where so many monuments were collocated together. On the other hand, as in some of our older graveyards, regularity of arrangement at different points might be merely incidental, rather than part of a general outline definitely adhered to; but, whatever rule (if any) determined the
- ↑ As I write of remains on French territory, it may be well to state that I do not use the word cromlech as applied in France, but according to its signification in England, where it designates that which the French call a dolmen, namely, a flat slab raised as a table, so to say, upon other stones set on edge. It is scarcely necessary to add that, according to French nomenclature, a cromlech means a circle of upright stones with or without another ortholith in the centre.
