In 1766 Nevil Maskelyne, fifth Astronomer Royal, compiled and published the first edition of The Nautical Almanac and Astronomical Ephemeris, designed to make practicable the determination of longitude at sea by the method of lunar distances. To mark the occasion of the two-hundredth anniversary of this event the staff of H.M. Nautical Almanac Office presents an objective assessment of the method of determining longitude by the measurement of lunar distances, in terms of modern knowledge, outlook, notation and computing facilities.

Reference to contemporary publications has been deliberately avoided; the story of the development of methods and techniques for the observation and reduction of lunar distances requires, and deserves, a separate study.

Dr. David Lewis, after completing the single-handed transatlantic race in 1964, sailed his catamaran Rehu-Moana through the Magellan Strait to Valparaiso and on to Easter Island and Tahiti. From Tahiti he sailed to New Zealand.

In an earlier paper published in the Journal (17, 278) Dr. Lewis reviewed the background to and evidence for various theories about early Polynesian navigation. One of his conclusions was that they must have been able to determine both course and, in effect, latitude with sufficient accuracy to make a landfall within the area of known groups of islands and so to proceed to any intended destination. This paper aroused considerable interest and was the subject of numerous contributions to the Journal, in particular from Andrew Sharp (18, 244) and Brett Hilder (18, 246).

In a series of communications written on board during the voyage to Tahiti, Dr. Lewis discusses the objections raised to some of his hypotheses. These cannot be dealt with here. On the voyage from Tahiti to New Zealand, via the Cook Islands, however, Dr. Lewis attempted to demonstrate, by navigating without instruments, that the methods used by the ancient Maoris were sufficiently accurate to make the voyages attributed to them navigationally feasible. It is this voyage that is related here.

In the first part of this series of papers an outline was given of the approach made at the Royal Aircraft Establishment to the problems of estimating collision risk and of specifying the quality of navigation needed to make separation standards safe. It was stressed that estimates should be based on intensive observation of flying errors, rather than on speculative theories, and that it is more feasible to develop ‘upper limit’ estimating techniques than those which purport to give the exact risk. In summary, a list of seven ‘requirements’ was given, as a reminder of the essential principles which can so easily be overlooked in the piecemeal task of relating separation standards to collision risk.

The purpose of this paper is to show some of the theoretical techniques which have been developed at R.A.E. to satisfy five of these requirements. (The remaining two do not call for special techniques and will be dealt with when practical applications are described in Part III.) The paper contains three Appendixes, dealing with the frequency of losing separation in one dimension, the computation of P's from the assumed tail shapes, and the treatment of relative errors. These are not included here but will appear in the off-printed version which may be obtained from the Royal Aircraft Establishment.

Both this paper and the paper that follows by Mr. Attwooll are crown copyright and are reproduced with the permission of H.M. Stationery Office.

In the first of this series of papers, the general principles of costing the effect of air traffic control deviation were discussed. This paper discusses the application of these principles to a model route structure for North Atlantic subsonic jet traffic. It is shown how the, design of an optimum system is related to cost considerations and how the cost depends on variations in traffic intensity and on the level of separation standards in all three dimensions. The work on which the paper is based was carried out at the Royal Aircraft Establishment.

The basic parameters leading to aircraft navigational error are combined statistically and are used to evaluate the uncorrected deviation (wander) to be expected in both subsonic and supersonic transport aircraft operating in a typical trans-oceanic environment. The size of the across-track and along-track wander before any form of corrective control is applied indicates the physical size of the A.T.C. block of airspace necessary to contain the aircraft. The variation of block size with the forward prediction period used by A.T.C. is presented graphically. The effect of improvement in the basic sources of error is demonstrated and it is seen that for the S.S.T. era attention must be turned to effecting improvements in aircraft heading and airspeed holding capabilities.

In America and Great Britain people are on the whole well acquainted with the picturesque circumstances surrounding the discovery of the ‘line of position’ by the Merchant Navy captain, Thomas Sumner. On the other hand, less has been written about the personality of Admiral Marcq de St. Hilaire, and about his work which led him to propose the comprehensive and exact method, now universal, which bears his name. My intention, therefore, is not to go back over the theory or the practice, nor even the value of this system of position finding by comparing points. It is just a question of making this brilliant naval officer better known, his upbringing, character, and the historic circumstances of the perfect achievement of which he is the immortal author.

This paper examines the requirement for a new type of display for the pilot navigator of the S.S.T. (supersonic transport). The author holds that inductive reasoning, which is the basis of manual navigation techniques, should not be dispensed with and the benefits of automatic systems will not be achieved until a similar awareness of the processes going on about the navigator can be reproduced. A well-designed display system might have this potential.

The ideas behind this paper were first described at the Annual Meeting of the American Institute of Navigation in Long Beach, California, in June 1965 and further developed at a meeting of the Deutsche Gesellschaft für Ortung und Navigation held at Munich in August of 1965.

This paper gives an account of a study carried out by the Institut Français de Navigation for the Administration de la Marine Marchande. The object of the study was to investigate the possibilities of introducing automation into the navigation of merchant ships. In addition to examining projects already underway in France and other countries, the Institut Français de Navigation sought the cooperation of numerous companies working in the electronics and computer fields and notably the Compagnie Générale d'Automatisme (C.G.A.), Compagnie Européenne d'Automatisme Électronique (C.A.E.) and Compagnie Générale de T.S.F. (C.S.F.).

Only the main features of this study are described here. Professor Hugon is Secretary-General of the French Institute.

Weather Radars. The presence of heavy rain and hailstones in cumulo-nimbus clouds is a natural phenomenon which is of vital importance to people who fly in aircraft. With a suitable radar fitted in the nose of an aircraft, this precipitation can be detected and the information presented to the pilot so that he can steer the aircraft to avoid the dangerous turbulent conditions associated with cumulo-nimbus clouds. This is, of course, the basis of weather radars which have been used for many years, and are now regarded as more or less essential for civil aircraft.

Progressive changes in the relationship of aircraft navigation and Air Traffic Control have produced a disturbing discrepancy between terrain clearance theory and practice. In theory, terrain clearance is assured by the aircraft adhering to certain specific and well-known traditional techniques; but in practice these are now often allowed to go by default, due to pressure of the A.T.C. environment. The purpose of this paper is to illuminate the exact nature of the problem and to suggest that clarification is overdue.

This report may be regarded as a continuation of that which appeared in Volume 16, No. 1, page 47 of this Journal. The results are based on precisely the same premises and are shown in the graphs as five-year averages. The graphs are drawn somewhat differently but the scales adopted illustrate the trends rather more clearly. As they cover the period of the last assessment as well as the subsequent three years, it will save referring back if the premises are restated.

Already many years ago Captain A. Vreugdenhil proposed the use of visual aids to supplement the sound signals used to indicate changes in heading. His idea of using visual aids was based on the erratic behaviour of sound and the consequent possibility of misinterpretation of sound signals.

Choosing between two possible systems, the directional illuminated arrow and, for example, flashing lights, he preferred and constructed an illuminated arrow, visible as such from a distance of approximately 1½—2 miles. He was against the use of synchronous flashing lights, pointing out the large difference in speed of propagation between sound and light signals and consequent possibility of errors when different sound signals are given in quick succession.

Commander Clissold (19, 109) has produced a very well considered argument to demonstrate that the Steering and Sailing Rules are inadequate for present-day use and that in the future they are likely to fall well short of traffic requirements.

I agree entirely with this assessment and I agree in general with his proposed solution. His suggested Rule requires action to be taken by both parties to every encounter where there is risk of collision, and the most important objection to this is that it implies dual responsibility for manœuvre. This is something which many people consider highly undesirable despite the fact that it apparently succeeds in the case of Rule 18, and to proceed directly from the present Rules to Commander Clissold's Rules would, I believe, be too large a step to be generally acceptable.

In my recent paper I proposed a set of manœuvring rules intended to solve the operational aspect of the collision problem. These rules are simple, but they are different if the bearing is lesser or greater than 90°. The object of this paper is to present a simpler rule valid for every bearing between 0° and 180°.

The Pair Rule. Except for the singular situation of both vessels meeting head-on or head-to-stern, every encounter involving danger of collision is characterized by the fact that the heading of both vessels is on the same side of the sight line. Close-quarter situations also involve, in general, those standing head-on, except for some situations with the heading on opposite sides of the sight line when one or both of them heads close to this line. Consequently, the evading manœuvre must be such that both headings are on opposite sides of the sight line and opening enough to ensure a safe passing.

The January issue of the Journal (pp. 124–5) contained a summary of Professor E. G. R. Taylor's views on the Vinland Map. To the courtesy of the Editor I owe the opportunity both to read this note in proof and to comment on it in the present issue. I have not seen Professor Taylor's complete study nor her drawings. She expressed some of her misgivings to me in correspondence after (in 1962) seeing a photocopy of the map and reading in typescript my study of the map, which is included in the recently published monograph.