+-----------+-------+--------+----------+----------+----+ | | Height| | | | | | | of | | | Distance | | | | Tower | | | of | | | | above | | |centre of | | | | first | | | gravity | | | | entire|Contents| Diameter | in feet | H | | |course.| of | at | at |from Base.| - | |Lighthouse.| (H) | Tower. |Base.|Top.| (G) | G | +-----------+-------+--------+-----+----+----------+----+ |Eddystone, | 68 | 13,343 | 26 | 15 | 15·92 |4·27| |Bell Rock, | 100 | 28,530 | 42 | 15 | 23·59 |4·24| |Skerryvore,| 138·5 | 58,580 | 42 | 16 | 34·95 |3·96| +-----------+-------+--------+-----+----+----------+----+ I come now to notice the few subordinate points in which the design of the Skerryvore Tower may be regarded as differing from those of the Eddystone and the Bell Rock. In glancing at the contrasted figures of the three buildings, it will be at once observed that the outline of the Skerryvore approaches more nearly to that of a conic frustum than the other two. To the adoption of this form, various considerations induced me; and these I shall very briefly detail. In the first place, it seemed to me that, in both the Bell Rock and the Eddystone, the thickness of the walls had been reduced to the lowest limits of safety towards the top; and the effects of the sea and wind acting upon a heavy cornice, cause a degree of tremor which I felt satisfied would not occur in a building with thicker walls. The effect of thickening the walls at the top, is, of course, _cæteris paribus_, to diminish the projection of the base, and thus to produce less concavity of figure, and consequently a nearer approximation to the contour of a conic frustum. I have already stated, that this excess of the bottom radius over that of the top, is in the Skerryvore Tower 13 feet, and that the height of the shaft is 120·25 feet. The quotient resulting from the division of the height by the excess of bottom radius over that at the top is 9·27; and, if the figure had been conical, this number would have given a measure of the slope of the walls throughout. There can be little doubt that the more nearly we approach to the perpendicular, the more fully do the stones at the base receive the effect of the pressure of the superincumbent mass as a means of retaining them in their places, and the more perfectly does this pressure act as a bond of union among the parts of the Tower. This consideration materially weighed with me in making a more near approach to the conic frustum, which, next to the perpendicular wall, must, other circumstances being equal, possess the property of pressing the mass below with a greater weight, and in a more advantageous manner, than a curved outline in which the stones at the base are necessarily farther removed from