So far the highly unstable phase of luminous blue variables (LBVs) has not been understood well. It is still uncertain why and which massive stars enter this phase. Investigating the variabilities by looking for a possible regular or even (semi-)periodic behaviour could give a hint at the underlying mechanism for these variations and might answer the question of where these variabilities originate. Finding out more about the LBV phase also means understanding massive stars better in general, which have (e.g. by enriching the ISM with heavy elements, providing ionising radiation and kinetic energy) a strong and significant influence on the ISM, hence also on their host galaxy. Photometric and spectroscopic data were taken for the LBV Var C in M33 to investigate its recent status. In addition, scanned historic plates, archival data, and data from the literature were gathered to trace Var C's behaviour in the past. Its long-term variability and periodicity was investigated. Our investigation of the variability indicates possible (semi-)periodic behaviour with a period of 42.3 years for Var C. That Var C's light curve covers a time span of more than 100 years means that more than two full periods of the cycle are visible. The critical historic maximum around 1905 is less strong but discernible even with the currently rare historic data. The semi-periodic and secular structure of the light curve is similar to the one of LMC R71. Both light curves hint at a new aspect in the evolution of LBVs.

An increasing number of non-terminal eruptions are being found in the numerous surveys for optical transients. Very little is known about these giant eruptions, their progenitors and their evolutionary state. A greatly improved census of the likely progenitor class, including the most luminous evolved stars, the Luminous Blue Varaibles (LBVs), and the warm and cool hypergiants is now needed for a complete picture of the final pre-SN stages of very massive stars. We have begun a survey of the evolved and un stable luminous star populations in several nearby resolved galaxies. In this second paper on M31 and M33, we review the spectral characteristics, spectral energy distributions, circumstellar ejecta, and evidence for mass loss for 82 luminous and variable stars.We show that many of these stars have warm circumstellar dust including several of the Fe II emission line stars, but conclude that the confirmed LBVs in M31 and M33 do not. The confirmed LBVs have relatively low wind speeds even in their hot, quiescent or visual minimum state compared to the B-type supergiants and Of/WN stars which they spectroscopically resemble. The nature of the Fe II emis sion line stars and their relation to the LBV state remains uncertain, but some have properties in common with the warm hypergiants and the sgB[e] stars. Several individual stars are discussed in detail. We identify three possible candidate LBVs and three additional post-red supergiant candidates. We suggest that M33-013406.63 (UIT301,B416) is not an LBV/S Dor variable, but is a very luminous late O-type supergiant and one of the most luminous stars or pair of stars in M33.

We discuss the spectrum of Var C in M33 obtained just before the onset of its current brightening and recent spectra during its present "eruption" or optically thick wind stage. These spectra illustrate the typical LBV transition in apparent spectral type or temperature that characterizes the classical LBV or S Dor-type variability. LBVs are known to have slow, dense winds during their maximum phase. Interestingly, Var C had a slow wind even during its hot, quiescent stage in comparison with the normal hot supergiants with similar temperatures. Its outflow or wind speeds also show very little change between these two states.

The progenitors of Type IIP supernovae have an apparent upper limit to their initial masses of about 20 solar masses, suggesting that the most massive red supergiants evolve to warmer temperatures before their terminal explosion. But very few post-red supergiants are known. We have identified a small group of luminous stars in M31 and M33 that are candidates for post-red supergiant evolution. These stars have A -- F-type supergiant absorption line spectra and strong hydrogen emission. Their spectra are also distinguished by the Ca II triplet and [Ca II] doublet in emission formed in a low density circumstellar environment. They all have significant near- and mid-infrared excess radiation due to free-free emission and thermal emission from dust. We estimate the amount of mass they have shed and discuss their wind parameters and mass loss rates which range from a few times$ 10^-6 to 10^-4 solar masses/yr. On an HR Diagram, these stars will overlap the region of the LBVs at maximum light, however the warm hypergiants are not LBVs. Their non-spherical winds are not optically thick and they have not exhibited any significant variability. We suggest, however, that the warm hypergiants may be the progenitors of the "less luminous" LBVs such as R71 and even SN1987A.