There is a long history behind the bubble sextant and it is a fascinating subject in its own right.
Before the advent of electronic navigation aids like Beacons, Consol, Gee and Radar, let alone GPS, the only way for an aircraft to navigate long distances without sight of the ground or by using rather risky dead reconing, was celestial navigation. However. There is one big problem. To measure the Altitude (Angle above the Horizon) of a celestial object, a navigator with a conventional marine sextant has to know where the horizon is. This is no easy feat in an aerial vehicle, be it ballon or aircraft. The horizon is usually very difficult to see and may well be below clouds or indistinct. Most Marine sextants were used early morning or late eveing when stars could be seen as well as the horizon.
Some success had been achieved on the Airships using marine sextants, sometimes with bubble levels attached. However. To use such an instrument was fiendishly difficult and they never really caught on. When aviators attempted to use conventional sextants they usually descended to an altitude as low as safety would permit. They were then able to use them in a manner similar to a marine navigator, making allowance for height of eye etc. There were obvious dangers in operating in this manner.
In the twenties many people throughout the world were employed trying to develop a practical bubble sextant.
At the Royal Aircraft Establishment, a Mr. Booth came up with a clever method of using a circular or "Bulls Eye" bubble chamber with tranparent top and bottom. One could then use this bubble with suitable optics to sight through the bubble at the object of interest. The RAE sextant was the resultant instrument and was known as the Mk V. Not long after, PF Everitt of Hughes and Son developed and produced the RAFs Mk VIII sextant. This was one of the first really practical bubble sextants and a similar design appeared in America as the National Bureau of Standards model and in Germany an almost identical instrument was produced by the firm of Plath in Hamburg. The Japanese Navy purchased several copies of the American sextant and made identical copies. These were later to be used in the attack on Pearl Harbour.
The firm of Henry Hughes and Son, whilst busy producing the Mk VIII had their chief designer P.F. Everitt investigate improvements. He came up with a much improved version. This was called the Mk IX, unsurprisingly. Several patents for it were granted in the following years. When you look at the design of this instrument you can see why it is streets ahead of any of the others. It is a relatively simple mechanism and is inherantly both rugged and accurate. It looks somewhat strange but almost every aspect of it has been thought out very thoroughly.
I hope to expand on this aspect of the bubble sextants when I get more time.
Early tests with the much improved "Booth" sextants initially produced mediocre results. It was realised by Francis Chichester, amongst others, that the problem was that an aircraft was accelerating in all three planes even though the aircrew might not be aware of it. The effect being to displace the bubble level away from true vertical. Francis Chichester also realised that if a series of sights were taken and the results averaged then a respectable result could be obtained. This is why most bubble sextants have some form of averaging device fitted. The RAF had a training aid to familierise trainee Navigators with the movements of aircraft. As a matter of interest the movements of the training aid are copied from those of a Sunderland flying boat.