In several studies of the Spanish and Dutch voyages in the Pacific in the sixteenth and seventeenth centuries it has been suggested that an apparent consistency in errors of latitude determined by the navigator in question can be used as evidence in the identification of island discoveries, and can be regarded as confirming the identification of one or the rejection of another possibility. I am extremely doubtful about the validity of using such apparent consistency of error, even as secondary evidence combined with other indications, and have no doubts about its invalidity as primary evidence.

Andrew Sharp, for instance, when considering the identification of the Los Bajos de la Candelaria of Mendaña's voyage of 1568 in his The Discovery of the Pacific Islands, remarks that whilst both Ontong Java and Roncador Reef in the Northern Solomons are possible identifications: ‘The steady southerly error in Gallego's latitudes … points to Ontong Java’.

It may be unexpected to some to find that among all the unsolved problems in the history of the magnetic compass there is a shortage of information on the form and arrangement of binnacles during the eighteenth and first part of the nineteenth centuries. The cupboard type of binnacle designed to take one or two steering compasses, according to the size of the ship, is pretty well documented for it is described in most of the books on navigation published in the French and English languages from the middle of the seventeenth century. There are also drawings in some books and there is a watercolour in the National Maritime Museum which shows the steering binnacle on the deck of His Majesty's Ship Deal Castle about 1775. This is however the only deck scene of the period known to me which does show a binnacle and very few binnacles appear in contemporary ship models.

At present there is no means of measuring the external temperature directly from a fast aircraft. This temperature is calculated by using the formula To = Ts/(1 + 0·2M2), using the mach number M given by the static and dynamic measurements of pressures, and the ram temperature Ts, measured by a resistance or thermocouple element. At high altitudes and when M is high, precision ranges from 1° to 4°, and the time constant from 3 to 4 seconds.

The Sun-Star Chronometer described in this note has been developed by Raymond Nardin of Switzerland and Oscar E. Batori of the United States; it is manufactured by Ulysse Nardin and produced by the Batori Computer Company Inc. of New York. It contains interesting features which merit description.

When the Offshore Powerboat Race was first held in 1961, navigators made the painful discovery that conventional methods of coastal pilotage were useless in a comparatively light, high-performance powerboat being driven to its limit.

Taking part in the race is like driving a Mini-Cooper over a frozen ploughed field in intermittent fog from London to York, the same distance as the course from Cowes to Torquay. Chart-work is out of the question. It is physically impossible to lay off a bearing or read a course. Furthermore, the figures and symbols are too small to be read, and, as often as not, the chart gets blown overboard. Radio bearings are unobtainable over the engine interference. A hand-bearing compass cannot be held steadily enough. Parallel rulers, dividers, pencils and the navigator's notebook end up in the bilges.

For a quarter of a century now we have struggled in a curiously impotent way with the problem of long-range navigation of civil aircraft and the associated problem of control of traffic in oceanic control areas. The deliberations of technical conferences of the International Air Transport Association alone during the past 15 years on the subject of Long-range Navigation of Civil Aircraft would fill a library.

The development of navigation has always been retarded by difficulties with the analysis of the operational requirement. It is, for example, inconceivable that if the operational requirement for astronomical navigation had been properly explained to eighteenth-century astronomers, they would have waited for practising seamen to invent the position line almost by accident. As neither the authors of Long-range Navigation of Civil Aircraft (D. E. Hampton and J. R. Mills. This Journal 17, 183) nor any of the speakers recorded in the subsequent discussion are directly involved with the operation of aircraft some remarks on the assumed requirements may be helpful.

The navigational problems presented by supersonic aircraft do not differ essentially from those with modern four-jet aircraft. Navigation will still entail three main tasks:

i. To ensure that the crew knows the position of the aircraft at any given moment, within a reasonable time, and with the necessary accuracy.

ii. To ensure that the aircraft is operated under the best conditions of route, speed and height for profitable operation.

iii. To ensure compliance with air traffic control.

‘Collision Avoidance by Discussion’ (see this Journal, p. 159) has been part of the radar controversy since it began. If it is assumed, as I think it must be, that no solution is acceptable which does not apply to multilateral situations, the complications inherent in such a proposal become apparent. A bilateral agreement, for example, on alterations of course will be based on the situation as seen at a given moment; this may change drastically within minutes as other ships' bilateral or monolateral decisions are put into effect. These may necessitate changes in intention and further confabulation, possibly now with others than the ship originally contacted, and so on.

One of the principal functions of air traffic control is to ensure that aircraft do not conflict, that is, do not approach each other sufficiently close to give rise to the possibility of collision. (Another function of air traffic control is to ensure that aircraft do not conflict with terrestrial objects, but this is outside the scope of this paper except in so far as such objects can be treated as aircraft with zero velocity.) The ‘possibility of collision’ is, of course, a matter for definition. It is often treated as an approach within 5 miles in plan and 500 or 1000 ft. in height (Fig. 1). The definition must depend upon the form of the aircraft track information available to the air traffic service—essentially flight plans updated by pilot reports—or the radar observations with which this paper is primarily concerned.

‘Surprise has sometimes been expressed by mathematicians who have been led by circumstances to look into the subject of nautical astronomy that no method is practised by sailors for reducing a longitude observation to the prime vertical in the same way that an observation for latitude is reduced to the meridian.’

This quotation forms the opening paragraph of the first of three papers by Mr. H. B. Goodwin, R.N., written in 1904–6 relating to the problem of reduction to the prime vertical. The following description of Goodwin's work in this connection may not be without interest in view of the recent interesting paper by D. H. Sadler in which the author describes the investigations into the same problem made by G. F. Cunningham.

The vast difference, to which Mr. Sadler refers, between proposing and translating into practice, a new navigational table, is exemplified in the case of Goodwin's P.V. Tables. All the intermediate steps between proposing and producing are described with utmost clarity in Goodwin's three papers.

Information on the shape of the ground can be required in the air for two quite different reasons. First, for visual fixing of position, that is to say, for ‘map-reading’; and second, for ensuring terrain clearance when flying on instruments. Obviously the second is of far greater importance than the first, yet unfortunately it is the first that, for historical reasons, has received the greater attention from cartographers; this paper will try to redress the balance, and indicate where action is needed at the present time to satisfy the second requirement.

(1) Through the above contribution by Captain Charles H. Cotter, my attention has been drawn to the fact that the essential principle of the ‘Rectofix Tables’ is not only quite old but has been incorporated in a set of published tables —the Position Line Star Tables (for fixing ship's position by reduction to meridian and prime vertical without logarithmic calculation) by H. B. Goodwin, R.N., J. D. Potter, London, 1906.

Understandably, there are considerable differences in methodology and presentation, but all the essential concepts underlying the method are incorporated in Goodwin's tables; the main improvement introduced by Dr. Cunningham (who was, of course, quite unaware of these tables) is the inclusion of tabulations for the azimuth (Journal 17, 17).

Although none of the colleagues to whom I showed my short article called my attention to the existence of Goodwin's tables, I must accept sole responsibility for not having made a more thorough search of the published literature on the subject. This omission does, however, serve to underline the general note in that article on the difficulty of assigning priority for particular methods.

In a recent contribution to Forum by Commandant L. Oudet (Journal, 17, 88) the problem of the rational form that marine radar displays should take has been raised. The old argument of ‘ship's head upwards’ or ‘north upwards’ may be resolved by a combination of the best of both systems without compromise to either school of thought.

Before starting on how this may be achieved a short digression on terminology should not be out of place. The advantages in the use of a ‘north upwards’ display, i.e. true bearings, lack of smear, accurate plotting and the ability to add ‘true motion’ do not stem fundamentally from the fact that north is at the top but because the picture is ‘compass stabilized’. The ‘north up’ condition is purely a matter of convenience. The ‘north upwards’ display would be better termed ‘compass stabilized north upwards’.

I have read with great interest the article entitled ‘Manned Space Flight Navigation Techniques’ by Major R. C. Henry in the October issue of the Journal.

It would appear that much of the imagined difficulty in interplanetary navigation is due to the fact that until now man has really been concerned with fixing his position and controlling his path on a spherical surface, the third-dimensional problem in air navigation has been one of altitude—i.e. definition of the surface concerned. Now, for the first time, we are involved in what at first sight appears to be a truly three-dimensional problem, and its solution appears difficult, partly because we are to be removed from our familiar graticule of earthly meridians and parallels.

If the problem be divided into two parts, an obvious solution presents itself. At sea, a ship must be on the surface of the water, and is mainly concerned with its position and course on that surface. An aircraft determines its surface of reference by barometric pressure (which by tradition is termed Pressure Altitude and expressed as a linear measure), and again is then concerned with a twodimensional problem on that surface.

If the day is not far off when communication satellites will lessen a navigator's dependence on conventional astronomical aids it is surely overdue for the deletion of obsolescent and superfluous items from the M.O.T. syllabus for candidates.

Of the three ‘standard’ formulae for finding a position line from an altitude of a celestial body (ex observer's meridian) only one, Marcq St. Hilaire, is in general use at sea today. The other two, Longitude by Chronometer and Latitude by Ex-meridian, depend for survival on the patronage of the Ministry of Transport and the Nautical Schools.

T‘HAT there is a general problem in avoiding high ground has been demonstrated somewhat catastrophically in recent years, but this paper deals with the descent phase of flight in a jet operation, where most of the incidents and accidents to scheduled international services have occurred. My own examination of these incidents and accidents (in some cases somewhat cursory) has led to the conclusion that in stating the problem in terms of ‘the avoidance of high ground’ there is the danger that we will miss the primary problem and indeed mislead the experts who might have little difficulty in providing us all with airborne radar designed for this particular purpose.