From 1928 to 1930 he worked under Edward Arthur Milne. In 1929, Milne had no problems left to ask his student to work on and appealed to Sydney Chapman, who proposed that they work on an article on which he was working that dealt with the Sun"s magnetic field Cowling found an error in the paper that invalidated Chapman"s results.
After Cowling"s doctorate, Chapman proposed that they work together.
In 1933 Cowling wrote an article, The magnetic field of sunspots. Joseph Larmor had worked in this area, arguing that sunspots regenerate themselves through a dynamo effect.
Cowling showed that Larmor"s proposed explanation was incorrect. His article assured him of a good reputation in the field of astrophysics.
During the 1930s, Cowling also worked on stellar structure involving radiation and convection, at the same time as Ludwig Biermann but independently of him.
He constructed a model of star with a convective core and radiative envelope, named the Cowling model by Chandrasekhar. He also studied magnetic fields within stars and classified the modes of non-radial oscillation of the body of a star, the basis of the field of helioseismology. Wilfried Schröder has discussed the relationship between Cowling, Chapman and Alfvén in a paper " Changes of theories of aurora borealis " (Gerlands Beiträge zur Geophysik, 1988).
Schröder has had long discussions with Cowling in this subject.
Cowling became an Assistant Lecturer at Swansea in 1933. He was subsequently a Lecturer at the University of Dundee (1937-1938) and the University of Manchester (1938-1945) before being appointed Professor at the University of Wales, Bangor (now Bangor University).
In 1948 Cowling was appointed Professor of Applied Mathematics at the University of Leeds in succession to Professor Selig Brodetsky. Cowling retired from his chair at Leeds in 1970 with the title Emeritus Professor.
Cowling was elected a Fellow of the Royal Society in March, 1947.
He was elected president of the Royal Astronomical Society from 1965 to 1967. Quarterly Journal of the Royal Astronomical Society 32 (1991) 201.