⁴⁰ They climbed to 11000 feet in order to get their morning speed check.Google Scholar

⁴¹ Not, however, the first aircraft. In 1911 the crew of the airship America abandoned ship and took to their lifeboat about 400 miles off the US east coast. They too were saved.Google Scholar

⁴² Kenneth MacKenzie Grieve like so many early AA practitioners was a naval officer who had been navigating officer of the seaplane carrier Campania during the First World War. He is quoted ‘I preferred to navigate chiefly by celestial observations and my position by the stars when picked up was practically correct. I used a cloud horizon instead of a sea horizon because the sea was hardly visible at any part of the time we were in the air.’Google Scholar

⁴³ Gago Coutinho devised a bubble sextant in 1917–19 but it is not clear to the writer that he understood the essential principle at that date.Google Scholar

⁴⁴ As the two-man crew must have been fully occupied taking sights and flying straight and level, the estimated position, and hence the stated errors, may not have been precise. The figures given are consistent with a random distribution, but the data are too meagre to assert such a distribution.Google Scholar

⁴⁵ A similar device (due to Waghorn) became a standard attachment to the Mk VIII sextant used in the RAF during the 1930s.Google Scholar

⁴⁶ Reproduced from this Journal, 6, 163. For a history of the study of acceleration errors in the UK from 1938 to the end of the piston-engined era see Hagger, A. J. (1952). The accuracy of bubble sextant observations.Google Scholar

This Journal, 5, 380. Some of the most important work on the subject in this period was done by Professor Plaskett, H. H. in conjunction with the Aeroplane and Armament Experimental Establishment.Google Scholar

⁴⁸ Very different results would emerge if the sights were taken from the rear gun-turret. In aircraft equipped with more than one astrodome inferior results were obtained at the dome most distant from the centre of gravity.Google Scholar

⁴⁹ Originally there was a clockwork drive and pick-up which carried every 2 seconds one sixtieth of the index glass setting to a totalizer. In later versions however Hughes introduced a cone, ball and cylinder type of integrator.Google Scholar

⁵⁰ Almost simultaneously the matter also appeared in the 1942 editions of Bennett's Complete Air Navigator and Weems’ Air Navigation. Bennett gives in Appendix VII a ‘ bubble bias correction’, Z where Z = X+Y. Table X gives the ‘ Coriolis or Geostrophic Deflection Correction’ in terms of ground speed and . The table correctly corresponds to the first term. Table Y give the ‘ Precession or Wander off course correction ’. The table correctly corresponds to the third term except that it is (erroneously) entered with TAS instead of Ground Speed. The author was clearly aware of the existence of the second term since he defines wander by reference to the great circle but he is quite silent on the question of how wander can be found. Since the Appendix is headed ‘ These tables are to be copied for use in the air’ the author must have been unaware of their imminent appearance in the Air Almanac. That was the year he was appointed AOC Pathfinder Force. Weems produces no tables but refers to the Air Almanac. The reference to the second term is more explicit and some attempt is made to address the question of how wander might be determined.Google Scholar

⁵¹Sadler, D. H. (1948). Altitude corrections for Coriolis and other accelerations. This Journal, 1, 22. Sadler wrote the note on which this paper is based in 1940. Col. T. L. Thurlow, USAAC, was the navigator of the Howard Hughes flight around the world in 1938. It seems in retrospect curious that Coriolis was not addressed by Professor Melvill Jones in his classic paper on acceleration errors [Melvill Jones, B. (1939). A Note on Sextant Observations in Flight. Aeronautical Research Council, 3874, NP 7, February].Google Scholar

⁵² For an interesting statistical analysis of errors of sights taken from an RAF Canberra, Aries IV, on a flight to the north pole in October 1954 see Parker, J. B. (1956). Acceleration errors in astro-navigation. This Journal, 9, 17.Google Scholar

⁵³ See for example Smart, D. M. (1931). Text Book on Spherical Astronomy. Cambridge University Press.Google Scholar

⁵⁴ Smart quotes Pulkovo Tables and Greenwich Tables.Google Scholar

⁵⁵ See Sadler, D. H. (1952). The Refraction Table in the Air Almanac. This Journal, 5, 223.Google Scholar

⁵⁶ Gago Coutinho at the time of the flight was a 53-year-old rear-admiral of the Portuguese navy. A few months earlier he had gone to Southampton to learn to fly well enough to relieve Cabral at the controls. Sacadura Cabral, 12 years his junior, was a commander in the Portuguese navy and his country's leading naval aviator. Cabral (like Alcock) was killed not long after his most memorable achievement but Coutinho (like Brown) lived into old age.Google Scholar

⁵⁷ A Gipsy Moth fitted with discarded damaged floats.Google Scholar

⁵⁸Chichester, F. (1964). The Lonely Sea and the Sky. Hodder & Stoughton.Google Scholar

⁵⁹ In 1827, when Sumner obtained his only Sun sight for days he realized that the ship was on the same ‘ parallel of equal altitude’ as Small's light and, the wind being favourable, sailed up his Sun line until he sighted the lighthouse. Perhaps Sumner has the prior claim.Google Scholar

⁶⁰ Reproduced from Bower, D. (1953). Navigation on recent RAF flights. This Journal, 6, 148.Google Scholar

⁶¹Richey, M. W. (1983). The province of the Institute. This Journal, 36, 1.Google Scholar

⁶² Among them the circularity of the area of probable position around the DR position and the very idea of the distinction between cases (a) and (b) based on values of 50 per cent probable error which were sometimes arbitrary to the point of caprice. If the areas do not overlap the data are not necessarily ‘bad’.Google Scholar

⁶³ Serson's artificial horizon stabilized by a spinning top. The true gyro did not appear for another century.Google Scholar

⁶⁴ In a lecture to the Royal Aeronautical Society in 1919, H. E. Wimperis, after addressing the problems inherent in gravity-sensing sextants stated, ‘gyrostatic methods of preserving the level are more attractive in theory, and thanks to the energy and ingenuity of inventors, there is good prospect of practical success being achieved ’.Google Scholar

⁶⁵ For a brief description see Fox, W. A. W. and Barnet, D. (1953). The vertical reference in aircraft. This Journal, 6, 161.Google Scholar