The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

July, 1913 iron, on the other hand, will acquire magnetism very quickly when laid in an N-S direction, but loses it. at once when turned at right angles to the earth's lines of force; its par- ticles show no tendency to re-arrange themselves under the shocks incidental to riveting the metal in place. To the combined effect of these two classes of iron, with the occasional complication of a constant quantity, all compass errors are due; and how- ever irregular the deviation table of a ship may appear, it always repre- sents the algebraic sum of two kinds of deviation, "semi-circular" and "quadrantal", each of which is per- fectly regular, symmetrical and easily separated from the other. Semi-Circular Deviation "Semi-circular" deviation is so called from the circumstance that it is east- erly on 16 consecutive compass head- ings and westerly throughout the other semi-circle, as the ship's head swings 360 degrees in azimuth. It is caused partly by the sub-permanent magnetism of the hull and partly by the transient magnetism induced in Wettical "soft iron such. as "masta davits, funnels, etc., by the vertical component of the earth's total force to north. When the compass is prin- cipally disturbed by the hull of the vessel the two points of zero semi- circular deviation correspond very closely to those points toward which the ship's head and stern lay while building, the deviation being easterly when that part of the ship which was south in building is east; when it is west. Thus an iron ship built heading north will show west- erly deviations on easterly headings and easterly deviations on westerly headings, there being no. deflection on north and south, where the forces of both ship and earth act in the Same straight line along the axis of the needle. Fig. 1 shows in diagram the effects of sub-permanent magnetism in a vessel built heading northeast. On this heading there can be no deflec- tion, for the north polarity on the Port bow acts in line with the earth's force; but since it repels the north end af the needle, the directive pull of the earth is diminished. (The ap- Parent inconsistency of this statement disappears when it is remembered that the north pole of the earth really Possesses south magnetism.) As the ship's head swings to east the north Polarity in the hull is drawn off to One side of the compass and repels It to the westward of magnetic north. On southeast this repelling force acts at right angles to the line along 'force westerly -- THE MARINE REVIEW which the needle tends to lay itself under the earth's pull, and the devia- tion reaches its westerly maximum. Heading south, the westerly deflection decreases, vanishing at southwest, where ship and earth are again act- ing in the same straight line. Here, however, they pull together, and the directive power of the needle is in- creased. As, the vessel swings. west- ward, the north polarity again drives away from magnetic north, this time sending it off to the eastward; and so the easterly deviation continues throughout 'the rest of the semi-circle. When plotted on the ordinary Napier's diagram, easterly deviations to the right and westerly to the left of the graduated center line, these deviations will show a pure semi-circular curve as in Pies 2 The deviations caused by vertical "soft" iron are equally simple. The vertical component of the earth's total (in north latitudes) induces north polarity in the lower and south polarity in the upper ends of all verti- cal and inclined elements. The com- bined effect, therefore, is polarity of a certain kind located at some fixed but often imaginary point in the vessel; and for this reason, like the fixed polarity of the sub-permanent mag- netism in the hull, it must cause semi-circular errors. nent magnetism, however, is a con- stant wherever the ship may be cruis- ing, and the sine of its deviation is always. proportional to that of the azimuth of the ship's head. The mag- netism induced in vertical "soft" iron, on the other hand, varies, the sine of its deviation changing as the tangent of the dip of a freely suspended needle, and for this reason must be different for different latitudes. Deviations Caused by Vertical Soft Iron - "Quadrantal" deviation is caused by the transient magnetism induced in horizontal "soft" iron, and derives its name .from the remarkable fact that it is always zero on the four cardinal points, shows maxima on the intercardinals and is alternately easterly and westerly in the successive quadrants. Due to horizontal induc- tion only, it does not change with the latitude, and is proportional, or more exactly its sine is proportional, to "that of double the azimuth of the ship's head measured from a point halfway between magnetic north and the direction of the disturbed needle. These deviations, as found in practice on shipboard, are always easterly in the first and third quadrants; wester- ly in the second and fourth. From this it is evident that the total dis- turbing cause seems principally due to deck beams, and may be represent- The sub-perma- 259 ed by a continuous thwartship bar of "soft" iron passing through the com- pass, as shown in Fig. 3. While the vessel heads north or south the bar is at right angles to the magnetic meridian, no magnetism is induced and there is no deviation. On the east and west headings, while the induced magnetism is maximum, it acts in line with the earth and there is no deflection. Between "north and east, induced north polarity on the port beam drives the needle to the east- ward; between east and south, to the westward; from south to west, to the eastward again, and in like manner, to the westward of north in the fourth quadrant. Laid down on a Napier's diagram as before, these deviations would show the quadrant curve in Fig. 4. Quadrantal Deviation Now it happens that a non-contin- uous, horizontal, thwartship "soft" iron bar also causes quadrantal deviation, but it is easterly in the second and fourth quadrants and westerly in the firse and third. © In Big. 5, -assthe vessel swings from north toward east it is the south end of the northerly half of the bar that acts on the com- pass, and the needle must, therefore, be drawn to the westward instead of being driven to the eastward as be- fore. In like manner the signs of all the deviations will be reversed; the curve in Fig. 6 can be made to exact- ly cancel that of Fig. 4, and thus the remedy for quadrantal deviation sug- gests itself. It has been stated that while the semi-circular errors due to vertical "soft" iron vary with the latitude, those attributed to horizontal "soft" iron do not, yet they both depend on a force which itself is constantly changing as the ship moves from equator to pole. The explanation is that in high latitudes where the earth's force is chiefly dip the vertical iron receives a large amount of mag- netism and exerts it accordingly. The horizontal iron is weakly mag- netized and the horizontal pull on (directive force of) the needle is weak in proportion to the magnetism in- duced in the ship. At the magnetic equator, however, there is no dip and the vertical iron can get no magnet- ism to exert, while the increased power induced in horizontal "soft" iron has to deal with a needle whose directive force and resistance to its influence has increased exactly as much as its own disturbing powers. "Constant" deviations are rare, and where found are generally more ap- parent than real. They asually re- sult from an error in the assumed