the slopes of the mountains overlooking the vale of Glen Roy are marked by
narrow terraces or parallel roads, which sweep round the shoulders of the
hills with "undeviating horizontality." These roads are described by Sir
Archibald Geikie as having long been "a subject of wonderment and legendary
story among the Highlanders, and for so many years a source of sore
perplexity among men of science." (517/2. "The Scenery of Scotland,"
1887, page 266.) In Glen Roy itself there are three distinct shelves or
terraces, and the mountain sides of the valley of the Spean and other glens
bear traces of these horizontal "roads."

The first important papers dealing with the origin of this striking
physical feature were those of MacCulloch (517/3. "Trans. Geol. Soc."
Volume IV., page 314, 1817.) and Sir Thomas Lauder Dick (517/4. "Trans. R.
Soc. Edinb." Volume IX., page 1, 1823.), in which the writers concluded
that the roads were the shore-lines of lakes which once filled the Lochaber
valleys. Towards the end of June 1838 Mr. Darwin devoted "eight good days"
(517/5. "Life and Letters," I., page 290.) to the examination of the
Lochaber district, and in the following year he communicated a paper to the
Royal Society of London, in which he attributed their origin to the action
of the sea, and regarded them as old sea beaches which had been raised to
their present level by a gradual elevation of the Lochaber district.

In 1840 Louis Agassiz and Buckland (517/6. "Edinb. New Phil. Journal,"
Volume XXXIII., page 236, 1842.) proposed the glacier-ice theory; they
described the valleys as having been filled with lakes dammed back by
glaciers which formed bars across the valleys of Glen Roy, Glen Spean, and
the other glens in which the hill-sides bear traces of old lake-margins.
Agassiz wrote in 1842: "When I visited the parallel roads of Glen Roy with
Dr. Buckland we were convinced that the glacial theory alone satisfied all
the exigencies of the phenomenon." (517/7. Ibid., page 236.)