Five seasons were spent in the erection of the beacon, so many were the obstacles which the workmen had to conquer, especially from the prevalence of rolling seas and stormy winds. In the fifth year, when the whole of the masonry had been completed, a November hurricane swept away the upper part of the structure. A modification of the original design was then adopted; and on the courses of masonry left intact by wind and waves six columns of cast iron were planted, terminating in a cast iron ball of three feet in diameter, which rose about twenty-five feet above the average sea-level. The whole was completed in the sixth season (September 1821).[41] [41] Stevenson’s “Account of the Bell Rock Lighthouse,” pp. 52-57. The Carr rock is about six miles north-north-east of the Isle of May lighthouse, and twelve miles south-west of the Bell Rock. In a recent pamphlet,[42] Mr. Thomas Stevenson, the engineer to the Board of Northern Lights—who has an hereditary as well as an individual claim to be heard on all matters of this kind—has suggested various modes of lighting beacons and buoys. As he observes, the importance of raising them to the rank of illuminated night-marks must be apparent to all who know anything of coast navigation; and he is certainly justified in thinking that the subject is worthy of more attention than has hitherto been given to it. [42] Stevenson, “Proposals for the Illumination of Beacons and Buoys” (ed. 1870). He speaks, in the first place, of _apparent_ or _borrowed lights_, where a ray is thrown on a buoy or beacon—as in the case of the Arnish Rock, referred to on page 274,—from a neighbouring lighthouse. The only other existing example of an apparent light is to be found at the harbour of Odessa, in the Black Sea. It was constructed in 1866, and is situated three hundred feet from the shore. Mr. Stevenson’s next suggestion applies to _dipping lights for sunk rocks_, where it would be difficult or impossible to erect a beacon for containing the necessary optical apparatus. Here he would so arrange the lamp and reflectors of the lighthouse as to dip vertically, and thus project a cone of rays upon the sea for a considerable area round the secret danger. On seeing the illuminated wave-space the mariner would alter his course, and give the sunken rock a “wide berth.” The other methods proposed by Mr. Stevenson are:— The conduction either of voltaic, magnetic, or frictional electricity, singly or combined, to the buoy or beacon, through wires, submarine, or, where practicable, suspended in the air, so as to produce a spark either with or without vacuum tubes, or by means of an electro-magnet and the deflagration of mercury. The conduction of gas from the shore in submarine pipes. Self-acting electrical apparatus, produced by the action of sea-water or otherwise at the beacon itself, so as to require no connection with the shore.