Magnetic control was a very early form of controlling electrons. The original Braun tube was magnetically controlled, and so was the first telephone relay of Robert von Lieben. Even the first successful attempt to obtain oscillations from a (mercury cathode) tube by Frederick Vreeland in 1908 was equipped with magnetic coils. Descriptions of magnetically controlled tubes were given by several authors [ [ [ [4] ] ] ] , e.g. in 1910 by H. Gerdien of the Siemens Co. and, two years later, a theory of electron motion in a coaxial diode, accompanied by measured results, was given by H. Greinacher. In 1918, A.W. Hull described the use of a coaxial magnetic field with a dynatron, and later he discussed the motion of electrons in a coaxial diode and gave the characteristic of plate current as a function of magnetic field strength [ [ [ [5] ] ] ]. He also proposed an oscillator using feedback via the magnetic coil. In
1924, E. Habann published a paper on the split-anode magnetron], which can be looked at as the first 'real' magnetron oscillator. Presumably, because of the lack of application of very high frequencies at that time, the German literature of the late 1920s seems to have forgotten about the Habann generator. The (at that time) famous books, e.g. of Barkhausen or Strecker, mention for the key-word 'magnetron' the Hull tube, and for highest frequency oscillators the retarding field tube following
Barkhausen-Kurz. The work on split-anode magnetrons during the later 1920s was mainly done in Japan by Okabe [22] and in the beginning of the 1930s by Megaw in England [21].Starting about 1933, the interest in high frequency oscillators, and thus in magnetrons too, arose again. The main interest of application, however, lay in the field of high frequency telephone links rather than in the beginning radar developments. A special problem was the modulation of magnetrons. One method was to install a control grid for amplitude modulation. Also discussed was 'Zeitmodulation', that is pulse modulation. But the problems of modulation remained severe: e.g. the commercial telephone link coded 'Rudolf', which first appeared in 1941, was initially equipped with a magnetron, but later redesigned for a triode transmitter [ [ [ [12] ] ] ]. Technological development led, on one hand, to high-power, water-cooled, split-anode magnetrons. On the other hand, there was the attempt to include the high frequency resonant elements into the vacuum vessel. In most cases these were Lecher type wires. The coupling to the outside was either capacitive or the lecher wires were led through the envelope. To the author's knowledge, cavities were used only in Barkhausen type tubes called 'Resotank'. When in 1943 the magnetron type CV 64 was found in a crashed British aircraft, the political restrictions for the development of cm-wave tubes vanished in Germany.