At present the total number of military and civil flights across the North Atlantic is about 130 per day counting both directions. Once aircraft are outside the systems of airways used in the high traffic density areas over each continent there are no fixed traffic lanes and aircraft plan their flights independently, generally following composite tracks to take maximum advantage of the wind distribution. Pilots are familiar with the natural tendency for tracks to diverge even when aircraft are being navigated via the same route but when the navigator in each aircraft is following an independent flight plan the separation of tracks will be much greater. Thus on the 2000-mile North Atlantic route the traffic density in real terms—say the number of aircraft per 10,000 square miles—is extremely low, and it is open to question whether the chance of two aircraft colliding in mid-Atlantic is not so remote as to be treated as an impossibility for practical purposes. This may sound heretical but if, as the writer suspects, the chance is so low as to make no significant difference to the overall risk of aircraft accidents, then there are other aspects of aircraft operation where the attention now being given to Atlantic traffic control might yield quicker dividends in the improvement of air safety.

There are still people who believe that only during the last four-hundred years have we known that the Earth is a globe. They were taught this at school when they were told the story of Christopher Columbus. Columbus, it was said, went to King John of Portugal with the proposal to reach the fabulously rich Cathay, situated in the Far East, by sailing across the Atlantic Ocean to the west. The king consulted a committee of mathematicians who rejected the idea with scorn. The Earth was round and flat like a trencher, they said, and not a globe as Columbus maintained. And some teachers added that not until the circumnavigation of Magellan's ship, Victoria, was there any proof of the shape of the Earth.

The pattern of port operation in fog which was established in Liverpool in 1948 when the Port Radar Station (Fig. 1)was opened, has not materially altered, except in the volume of shipping now using the service provided.

A straightforward shore-based radar has been shown to be a reliable and effective aid to pilots in bringing ships in from sea in fog through the port channel area into the berthing area, with or without the assistance of the ships' own radar. Once there, however, further movement now depends upon the visibility in relation to such factors as radar assistance, shipping density, the presence or absence of emergency anchorage, the nature of the berth, that is, floating stage, river wharf, or dock, and so on. Below a certain visibility, shipping hold-ups and increased risk of stranding for ships on the move may be expected, As an ultimate ideal, we may have in mind the integration of all available facilities and inventions so as to allow shipping to carry out all its normal activities in the worst of fogs, with no less speed and safety than is customary in clear weather. But this is rather idealistic, and I would like to introduce a criterion, suggested by Mr. Eric Parker, which he terms the ‘Critical Visibility’, and which is the visibility at which 50 per cent of ships will proceed no further than the berthing area. This statistical measure of the effect of fog is one which will give a real indication of the value of any improvements. It will of course be different in different ports, but if say it is half-a-mile with no shore radar assistance, the introduction of shore radar might reduce it to quarter-of-a-mile, and in a few years time it might be reduced by improved aids to say a hundred yards. Eventually it would be at such a value that annual delays to shipping would be of negligible importance.

The advent of true-motion radar has been a great step forward in the problem of allowing a ship to operate safely in congested waterways. Harbour radar, however, is still a requirement for two reasons; first of all it can well provide far more immediate information than is available from true-motion radar. Fig. 1 emphasizes this by showing how much additional information can only be available directly from a static site. Consequently it allows shore authorities to be advised of what is happening in conditions of bad visibility where they would otherwise be devoid of most information. This raises the problem of how the information is to be used.

The main problem in polar navigation is one of direction, and the problem has two facets. First we have to measure direction and then we have to describe it. Generally, the seaman has a gyro-compass by which he can measure the heading of his ship. As you know, the gyro-compass is simply a gyro which keeps pointing towards the pole. If we have a gyro close to the north pole and pointing north, it will continue to point in the same direction however the ship turns beneath it. So also will it continue to point in the same direction if the Earth turns beneath it. In twelve hours it will be pointing south instead of north.

Captain Scott went to the pole over the Ross Barrier, up the Beardmore Glacier (Fig. 1) which is about 100 miles long, and then on to the south pole, coming out by the same route. The great Norwegian explorer Amundsen went in from the other side over the shelf, up one of the glaciers and on to the pole. Admiral Byrd flew in and out again in 1929. Apart from these expeditions very little has been done in Antarctica and nearly all of it, apart from the fringe, is untouched.

The operation of helicopters in all weathers is an essential development if full advantage is to be taken of this class of vertical take-off aircraft. Helicopters must operate in weather minima at least as low as the current fixed-wing limits of 300 yards visibility and 300 ft. cloudbase, and should approach the limits approached by surface transport which continues to operate down to visibilities of the order of 10 to 20 yards. Present experience suggests that it should be possible to conduct regular operations down to 200 yards visibility and 200 ft. cloudbase provided the necessary aids are available. As experience is gained, these limits might be reduced to 150 yards and 150 ft. However, as the pilot needs time to change over to visual guidance before landing, it does not seem likely that these limits could be much reduced before the development and acceptance by civil operators of either completely automatic landing or some radical new landing aid.

The first equation displayed in Appendix A to Mr. J. E. D. Williams's paper (Vol. X, p. 44) should have read:

The object of this paper is to compare modern interception theory with the better known classical concepts. We shall examine the curve of pursuit, the classical straight-line interception and the modern theory of proportional navigation.

The mariner is, on the whole, a cautious creature. When he takes action, he likes to know within fairly narrow limits how much is necessary and where that much will take him in relation to all potentially dangerous objects. He has as much respect for the fire as for the frying pan.

When plotting a series of radar bearings to determine the other vessel's course and speed, any alteration of own course or speed during the plotting interval greatly complicates the construction of a relative plot. A continuous alteration would make it very difficult to construct a plot at all. As far as the other vessel is concerned, should she also have radar, your echo on her PPI will not follow a straight line and she will be able to draw no conclusions regarding your course and speed. There is therefore no doubt that, when using radar for anticollision purposes, alterations of course and speed should, as far as circumstances permit, be as few as possible and each should be bold. Let us consider what is ‘bold’, treating alterations of course and speed separately, and remembering that it is necessary for the other vessel, if she is using radar, to appreciate that you have altered course as early as possible after your action.

In clear weather, many of us have used the curve of pursuit to avoid a vessel converging from well out on the starboard bow. We have said ‘Starboard’ and followed that order with ‘Head straight at her and report when you have regained course’. Convenient and safe though this technique may be in clear weather it can hardly be recommended for use with radar in fog. Apart from anything else, the other ship may have radar and may also take action as the distance between the two ships decreases. The slowly changing nature of our course would make it well nigh impossible for own ship's radar plotter to determine what action the other ship was taking, and might easily lead to disaster.

A simple device to assist the graphical solution of problems involving relative motion is described, in the hope that it may contribute something to the interpretation of radar plotting.

Introduction. Basic requirements from a radar plot are:

• (a) the speed and course of the other ship;

• (b) an appreciation of the effect of a change of course or speed; and

• (c) a ready determination of the change of course or speed to produce a required effect.

It seems to me that the radar screen does not present collision avoiding information in the best form and that we should give some thought to the way in which we would like this information to be presented. Then we should ask the back-room boys to present it to us electronically. It must, of course, be remembered that a ship does not occupy a geometrical point and it will want to escape collision by more than the merest miss. Therefore we should build a margin of safety into the presentation.

In his Presidential Address at the Annual General Meeting of the Institute, Captain Majendie spent some time on demonstrating the restrictions imposed on free navigation by air traffic control. Whilst there is much truth in his statement the matter does not end there and I believe that those who study his address in retrospect should remember that the marked inflexibility of economic operation in the vertical plane of the modern jet aircraft and, to a lesser extent, the turbo-propeller driven aircraft, imposes considerable restrictions on air traffic control! In fact, the designers of modern aircraft and engines seem to be doing their best to deprive air traffic of its unique asset—the third dimension.

Last summer I sailed in R.M.S. Newfoundland from Boston to Liverpool and took the opportunity of keeping up a navigational plot using a minimum of charts and instruments. The sextant I used was a box sextant and it occurs to me that some description of this instrument may be of interest to navigators. Box sextants are lightweight and usually have a sling case which enables them to be carried or packed away easily. They are much less expensive than the standard sextant, and since they are precision-made instruments will last just as long. Good results can be obtained with them both in coastal navigation and offshore.

H.M.S. Dryad's memorandum (Journal, 9, 444) speaks of ‘only those tables required’. But what this word ‘required’ does or should mean could be debated ad infinitum. The main division comes as between a set of tables intended to be an instrument of practical navigation and one intended to be an instrument of technical education, and the dividing line between the two would be a very blurred and disputable one indeed. My own set of nautical tables came into existence because I was convinced at the time that a set of tables, ‘streamlined’ for sea use was urgently needed, and in so far as logarithmic methods may be still used I still consider that my first edition was a better instrument of navigation than any of the four later ones. But whereas I knew that I wanted such a set of tables for my own use, and was of the opinion that everybody else did (or should), I became aware that I was in a comparatively minute minority. In fact it became slowly and painfully borne in upon me that something like 90 per cent of all navigators would use all their sea-going lives the tables they were first taught on. This meant that I had to consult the schools and examining authorities and include all their academic needs, and so my streamlined volume soon began to develop a corporation. I did, of course, try again with my Four-Figure Navigation Tables, but that is another story.

Since the advent of radar, the need for correct interpretation has become only too clear from the record of so called radar-assisted collisions. The practical navigator has been shown various methods of noting PPI data and solving the problem of how best to avoid collision. We have the practical system of plotting, either with ship's head or true north upwards. We are now being offered some ingenious graphs and tables by Captain Wylie, Mr. Slater and others (Journal, 9, 161, 448). The navigator has an opportunity to study these methods and use them at sea, in clear weather, in conditions where he can assess the advantages of each method and also their accuracy.