Until now studies on the historical development of atomic spectroscopy have focused on three main aspects-its first applications as a method of chemical analysis, the formulation of spectral laws , and the rise of the old quantum theory. These developments of spectroscopy were based on the same assumption: the invariance of the atomic spectrum after fixing the chemical element . This paper shows that running alongside these lines of research there was another, no less important area of study based on (...) the negation of this assumption. The focus of these latter studies was the behaviour of atomic spectra as a function of different working conditions. English astrophysicist Sir Norman Lockyer played a central role in this research. While studying the effect of temperature on spectral lines and comparing stellar spectra with laboratory spectra, he discovered a new class of spectral lines . In an attempt to explain the origin of these lines, Lockyer, at the beginning of 1897 , hypothesized the 'dissociation' of the chemical elements into simplified forms of matter, named proto-elements. Lockyer's studies are mentioned by Thomson in the paper which introduces the corpuscle for the first time. In this paper we have tried to reconstruct the history of Lockyer's proto-elements and highlight the links between them and Thomson's corpuscle. Our analysis has shown that not only the famous studies on discharges in rarefied gases but also Lockyer's stellar spectroscopy played an important role in the work of J. J. Thomson, convincing him that the atom was complex and divisible. (shrink)

Historiography has pointed out that the time between the mid 1910s and the early 1930s can be considered a pivotal period in the history of stellar astrophysics. In those years, scholars like Saha and Eddington first applied atomic physics to astrophysics. Theoretical astrophysics was born. This led to the development of the first physically sound models for stellar interiors and atmospheres. These landmark achievements spurred scholars to elaborate theories for stellar evolutions, and in the following decades several astrophysicists focused on (...) this problem. The evolutionary role of red giants turned out to be the main issue. Those stars were initially assumed to be young ones going through the formation stage, but astrophysicists gradually realized that they were rather to be considered old, evolved stars. The solution of the giant stars issue required a couple of decades: it was not until the mid 1950s that a satisfactory explanation was obtained. This provides a detailed picture of the theories of stellar evolution from the 1930s to the 1950s and of the solution to the red giants problem, with special emphasis on how such a solution was made possible by a series of subsequent steps: the identification of changing chemical composition as a main evolutionary feature of a star, the inclusion of nuclear physics within the theoretical framework of stellar astrophysics, the recognition of the importance of inhomogeneities that settle within stars as nuclear processes go on. (shrink)