It would be interesting if we could know how Giraldus gathered these data. Probably they were pieced together from incidental observations of sailors or fishermen, for certainly no systematic investigation of tidal phenomena could well have been undertaken at Giraldus’ time.[886] It is typical of an immense amount of close and accurate knowledge that has always existed along with ignorance and superstition among the more humble workers of this world, knowledge that until recent years has but rarely found literary expression. OTHER MARINE PHENOMENA NOTED BY GIRALDUS Giraldus certainly was not always so fortunate in his discussion of marine phenomena. He taxes our credulity a little when he tells of a rock in Arklow harbor on one side of which the water rises while it is falling on the other,[887] though this may perhaps have resulted from some local play of currents and eddies. It is less easy to find an explanation of a story which he relates of a recession of the sea at “Crebonensus” (Proconnesus?) near Constantinople.[888] Here, during eight days at the time of the festival of St. Clement, the waters fell back in order to allow pilgrims to go to the saint’s shrine. This kind of miracle, to be sure, had the support of Biblical authority in the story of the parting of the waters of the Red Sea to permit the passing of the children of Israel; and we find a similar tale in the _Otia imperialia_,[889] where Gervase asserts that the Sea of Pamphylia was divided for Alexander the Great, because God wished to destroy the Persian kingdom by means of the Macedonian. The lake (or river) which in the legend surrounded the church of St. Thomas in India was also said to go dry at regular intervals to permit pilgrims to approach.[890] In his description of South Wales, Giraldus Cambrensis gives an account of marine encroachments on the land and perhaps of coastal subsidence. A great storm on this sandy coast laid bare a forest hitherto covered by the waters. Trunks of trees appeared with marks of the ax upon them, fresh as if cut only the day before. Giraldus was convinced that the marks dated from inconceivable antiquity, perhaps even from the time of the Flood.[891] The wood was overwhelmed, he said, by the constant and ever increasingly violent advance of the sea; and certainly it is well known in modern days that the waves long have been eating into the coast of Pembrokeshire and that the uncovering by storms of buried forests and stumps is a commonplace occurrence there. Perhaps we are justified in interpreting Giraldus’ remarks by assuming that the forest had not, as he states, previously been covered by water but more probably by marine sands or muds, which subsequently were removed from the stumps by storm waves. ST. BRANDAN AND THE SPIRIT OF THE SEA