It is when these winds prevail that they are most frequently found by the outward-bound vessels-between the latitudes of 40° and 50° S.

Having thus given some notion as to their general character, we may add some admonitions as to avoiding them in sailing where they may be supposed to exist.

The indications of an iceberg are:-1. A natural effulgence, or ice-blink, which frequently renders them visible at some distance, even in the darkest night. At a short distance this effulgence may appear like a white cloud, extending over, or nearly over, the vessel's masts.

2. A considerable decrease in the temperature of the water, as shown by the thermometer, in comparison with the heat of the adjacent sea, and with the air above.

3. The roaring of the sea at the base of a berg, which, except in a steamer with its paddles in action, may be heard by an attentive listener when afar off.

Capt. Weddell recommends that, with a free side-wind, an iceberg or ice island should be passed on the windward side; as by this means the loose ice, which always drifts farthest, is avoided. If the ship, too, be moving with some degree of rapidity, she can avoid these small pieces more readily, as she is then more obedient to her helm. The large ice islands are not the most dangerous to a ship in passing among them, as they can be more easily avoided; on the contrary, it is the small, broken, or detached pieces, level with the water's edge, which are the most mischievous; for, when the wind is high, it is almost impossible to distinguish them from the break of the sea, and yet these small pieces do as much injury to a vessel as large ones, by knocking a hole in her bottom.

Capt. Boulton says that dependence should not alone be placed on what is said to foretell the near approach of icebergs; viz., a white or luminous reflection in the atmosphere over them. This may sometimes occur; but, from strict observation, it is ascertained that it can be discovered only over those which are large and square-topped, besides being invariably covered with snow. The rugged icebergs, and those that have upset, never show themselves in that manner as far as observation enabled to decide. The safest and the best way to discover them is by keeping a good look-out; the eyes constantly tracing and retracing the dark line of the horizon, for ice will always make that part of the horizon, where it is, lighter. By adopting this method they were never mistaken; whereas, if they had been looking aloft, they would have run on many; at the same time one individual should be more particularly appointed to look out for the small pieces.

There is one use in the floating ice; that which is clear and transparent, without flaws or enclosed apertures which will contain salt water, will afford the purest and most delicious water in nature, and is a ready means of adding to the ship's stores. On the flat bergs or field ice pools of this fresh water are sometimes found beneath a scum of ice, and the water will be found perfectly wholeSome remarks on this are given in the Appendix.

some.

1276

CHAPTER XXXIX.

PASSAGES.

THERE is one misfortune attendant on the advancement of science, that by following out to minute particulars each special branch of it, the mind is more or less diverted from the simple first principles. The whole tendency of modern research, so indefatigably carried on in the present day, is to multiply the facts attendant on any department of physics; that, instead of the plain matter of fact with which our forefathers were content only to know, we have now such a multitude of phenomena to deal with, that the real question is often subsidiary, or lost sight of. Thus the variation of the compass, the range and hour of the tide, the direction of the progress of winds and currents, have been shown, by multiplied observation and discussion, to be, though more or less simple in the principles in which they originate, most complex in their action, and very different matters to comprehend to what in olden times they were understood to be. There is no department of practical science to which these remarks are more applicable than in that of navigating a ship. When the rude instruments, and the imperfect resources of the navigator in ancient times are considered, it is rather a matter of wonder that ships were conducted to distant countries at all. But yet a perusal of the accounts of the voyages, made with these means only, will soon convince us that passages were then made fully equal in rapidity to many now undertaken, and, of course, very much in advance of all that might be expected by judging from what knowledge they have left us. Perhaps it was the very imperfection of their means which led them to use diligence and forethought, which are now replaced by more scientific substitutes. One thing is certain, that the simplicity of the various branches of navigation, as then understood, allowed the mariner to comprehend them as a whole better than in the present day, when they are subdivided into such an infinity of minor details, and the great principles of the sphere were then better considered than now, when the universality of charts causes the surface of the earth to be rather viewed as a plane in hydrographical problems. What we would wish these remarks to tend to is, that in former times a principle in navigation, that of sailing on a great circle, was better understood and carried out than in the present day, when it is now being revived as a NEW subject; and, as the Pacific Ocean is certainly the peculiar sphere for carrying out this principle, we will be the more diffuse on the point.

Very early in the days of navigation, prior to those transatlantic voyages which led to the discovery of America, and the sea voyage to India, the principles of the art of great circle sailing were understood and promulgated. Sebastian Cabot alludes to it in 1495, and it is more than probable that Columbus, Magalhaens, and all the first great navigators were familiar with the subject.* It must

The first work, apparently, in which it is directly alluded to, is by Pedro Nunez, in 1537. Another is by Pedro de Medina, in 1545, in the Spanish language, but his system was erroneous,

be remembered, that at this time the principles of longitude were not understood or defined, and the charts of the period were merely the result of measurements made by dead reckoning or estimation, a method not in disuse until more than a century later. Almost all the early works give instructions for making these "sea-cardes," as well as the rude astronomical instruments, "cross-staffes," "" astrolabies," &c., which were then considered sufficient.

A most important epoch in the history of navigation now succeeds. Gerhard Mercator, in 1569, published a universal map constructed on the principle now known by his name. In this the meridians, as well as the latitude circles, are represented by parallel straight lines; and by augmenting both the latitude and longitude in the same proportion, the rhumbs, which in reality are curves or spirals, become represented on it by straight lines also. Mercator does not appear to have exactly comprehended the true principles of his projection. This was reserved for Edward Wright, an Englishman, who correctly described its nature in his work called "The Haven-finding Art, or the Way to find any Haven or Place at Sea, by the Latitude and Variation, 1599." Still there is no mention of longitude as an element of navigation; but this soon was understood, and consequently the simplicity of the sphere was lost sight of in the facilities given. by Mercator's or Wright's projections, and the ascertaining of the approximate longitude, although the theory of great circle sailing is of little use without longitude.

Without divesting the mind of the ideas implanted by the consideration of a plane chart, it is somewhat difficult to comprehend the exact nature and practical application of the great circle. It is by this method only that a ship can be directed to her destination as "the crow flies," or as if it were in sight, and the deviation from the systems usually adopted for convenience, from the charts, is greater, according as the distance between the two points is greater. It is most readily comprehended by observing how a thread stretched tightly over an artificial globe cuts the meridians and parallels.

The shortest distance between two points on the surface of a sphere is a portion of an arc of a circle which cuts these two points and would surround the sphere, having the same radius and centre as the sphere itself. The equator is such a great circle, thus named because it is the largest circle which can be drawn on the sphere. A meridian is also a great circle, and cuts the equator at right angles. The intersection of two or more of these meridians is at the North and South poles, 90° of arc distant from the equator. By sailing exactly on the equator, or on a meridian, are the only directions in which we shall find the compass to maintain exactly the same direction throughout a passage which shall be the shortest distance between any two points on the earth's surface. In a direct

and was corrected by Martine Cortes (or Curtis), whose work, “The Arte of Navigation," was soon after, in 1561, translated into English by Richard Eden, and was long the text-book of British seamen. Numerous other works, in which it is distinctly and correctly described, afterwards appeared, as one by Michael Coignet, of Antwerp, in 1581; an excellent work by Roderick Zamarano, in 1585, &c. That by this time it was thoroughly recognised is evident by a work by John Davis, published in August, 1594, called "The Seaman's Secrets, wherein is Taught the Three Kinds of Sayling-Horizontall, Paradoxall, and Sayling upon a Great Circle." It is also described in Richard Polter's "Pathway to Perfect Sayling," about the same time. After this it is found in most of the old works on navigation.

track on any other circle than a meridian or the equator, or due East or West, or North or South, the true bearing of the track will vary with each change of place.

This apparent anomaly will be cleared up if we consider what is the real nature of the angle termed the bearing of one point from another, as indicated by the compass or other means. East and West are terms referring to the horizon of a place; but these are only relative to the direction of the true meridian, or the North and South line of such place, and the East and West line cuts this meridian at right angles. But meridians are not parallel on the earth's surface they, though straight, meet at different angles at the poles. Therefore any straight line at right angles to one meridian will not, if continued, cut any other meridian at a right angle, because they are not parallel; but the angle will vary more or less from a right angle according to the distance these meridians are apart.

Now in Mercator's projection, and here is the difficulty, the meridians are all made parallel consequently a straight line intersecting one meridian at any angle will, if continued, intersect any other meridian at precisely the same angle. The straight intersecting line, on the surface of the plane in Mercator's chart, represents a very different thing on the spherical surface of the earth.

But if at any other angle it becomes a rhumb line, and this line transferred to the spherical surface becomes a spiral, and continued infinitely would encircle the globe, gradually approaching the pole, which it never reaches. Mathematically this curve, the loxodromic curve as it is called by the older writers, is one of very great complexity, but its simplicity, as practically applied to navigation, has caused it to supersede the apparently more difficult great circle problems.

A great circle may be defined as a circle which divides the earth into two hemispheres, using this last term without reference to its usual meaning, and necessarily may vary in inclination to every possible angle from the equator or any one meridian.

The equator, being a great circle, necessarily bisects every other great circle on the earth, whether at right angles to it, as the meridians, or at any inclination. A great circle, therefore, which is inclined to the equator, which must be the case with those which pass through two places, in different latitudes, passes through two points on the opposite side of the sphere, in directly the opposite latitude and longitude, as it must be bisected by the equator. And there are two points also of more importance in the calculations, and these are the points where the circle attains the greatest amount of divergence from the equator, or the maximum latitude attained in each hemisphere. These two points are called the vertices* of the great circle.

It follows that the arc of a great circle and the rhumb line differ most widely from each other in high latitudes, and between places nearly on the same parallels. In low latitudes the two curves nearly coincide. The difference, too, is not so great when the two places are on opposite sides of the equator, because the great circle and the rhumb line then intersect each other.

The Hydrographical Office have published a useful set of tables for facilitating great circle sailing, by Mr. Towson. The construction of these are dependent on the latitude of this vertex, and the angular distance from its meridian, or that which besets the circle at right angles, and with the equator divides it into four quadrants.

The ship, in sailing on a great circle, is always in a higher latitude than when sailing on a rhumb line; hence if both tracks coincide at their extremities, there must be a point in the great circle at which its distance from the rhumb line, measured on a meridian, is greater than anywhere else; this point is called by Lieutenant Raper the point of maximum separation in latitude. It is by means of this point, and the two extremities of the arc, that Lieutenant Raper proposes to lay down, roughly, the great circle course on a chart.

We cannot here give the working portion of the great circle problems; that must be left to books specially devoted to the subject. It will be found, on referring to them (as in Lieutenant Raper's "Navigation," 3rd edition, pp. 105— 113, and pp. 124-126, as well as Towson's "Tables" previously alluded to), that the immense labour, formerly attendant on the necessary calculations, is greatly simplified and reduced; still much is to be desired before it can be brought to that necessary simplicity to enable the mariner to combine this system with the numerous other considerations which bind him.

There is one advantage in the great circle track, and it is no mean one. The great circle, apparently a circuitous route on the chart, represents a shorter distance than the straight rhumb line. Therefore if a ship be navigated anywhere between the great circle arc and the rhumb, she will still shorten her track. And further, if she assumes a course as much higher in latitude as the great circle course appears to be, she will still not have to sail over a greater space than the rhumb line. This consideration opens a wide field for choice as to a proper parallel to sail upon.

There are very considerable difficulties and apparent contradictions to the usually received notions, in judging as to the best course. A mathematical formula may present the exact directions and extent of an arc of a great circle; but another point arises in its practical application, that is, where does it lead to throughout its course? into what latitudes, or into the neighbourhood of what islands or countries? And again, by assuming a course so very distinct from that which the rhumb course laid down on the Mercator chart, will it, by carrying the ship out of the trade-winds or equatorial currents, or the reverse, neutralize or reverse the advantage which its shorter distance will give?

If the seaman, instead of using the plane chart, could use the terrestrial globe to guide his course, all difficulty would vanish, and the subject of great circle sailing would become clear to the mind of every one. It is its apparent anomaly with that of Mercator's projection which constitutes all the difficulty.

For example:-From the entrance of the Strait of Juan de Fuca (lat. 481° N., lon. 124° W.), Guam, in the Ladrone Islands (lat. 13° N., lon. 1443° E.), bears about East, but the Strait of Juan de Fuca bears N. W. from Guam, and these ought to be the respective courses on starting from either position. By Mercator sailing the respective bearings are nearly E.S.E. and W.N.W., which courses, if maintained throughout the passage, will conduct a ship from one point to the other.

Again: From a position 30 miles South of the Diego Ramirez Islands, to the entrance of Cook's Strait, in New Zealand, the great circle course touches the antarctic circle in about lon. 117° W. Now it is manifest, upon reading the