We examine the photometric data for Fornax clusters, focussing our attention on their horizontal branch color distribution and, when available, on the RR Lyr variables fraction and period distribution. Based on our understanding of the HB morphology
in terms of varying helium content in the context of multiple stellar generations, we show that clusters F2, F3 and F5 must contain substantial fractions of second generation stars (~54-65%). On the basis of a simple chemical evolution model we show that the helium distribution in these clusters can be reproduced by models with cluster initial masses ranging from values equal to ~4 to ~10 times larger than the current masses. Models with a very short second generation star formation episode can also reproduce the observed helium distribution but require larger initial masses up to about twenty times the current mass. While the lower limit of this range of possible initial GC masses is consistent with those suggested by the observations of the low metallicity field stars, we also discuss the possibility that the metallicity scale of field stars (based on CaII triplet spectroscopy) and the metallicities derived for the clusters in Fornax may not be consistent with each other. The reproduction of the HB morphology in F2,F3,F5 requires two interesting hypotheses: 1) the first generation HB stars lie all at red colours. According to this interpretation, the low metallicity stars in the field of Fornax, populating the HB at colours bluer than the blue side ((V-I)o<=0.3 or (B-V)o<=0.2) of the RR Lyrs, should be second generation stars born in the clusters;a preliminary analysis of available colour surveys of Fornax field provides a fraction ~20% of blue HB stars, in the low metallicity range; 2) the mass loss from individual second generation red giants is a few percent of a solar mass larger than the mass loss from first generation stars.
Among the newly discovered features of multiple stellar populations in Globular Clusters, the cluster NGC 1851 harbours a double subgiant branch, that can be explained in terms of two stellar generations, only slightly differing in age, the younger o
ne having an increased total C+N+O abundance. Thanks to this difference in the chemistry, a fit can be made to the subgiant branches, roughly consistent with the C+N+O abundance variations already discovered two decades ago, and confirmed by recent spectroscopic data. We compute theoretical isochrones for the main sequence turnoff, by adopting four chemical mixtures for the opacities and nuclear reaction rates. The standard mixture has Z=10$^{-3}$ and [$alpha$/Fe]=0.4, the others have C+N+O respectively equal to 2, 3 and 5 times the standard mixture, according to the element abundance distribution described in the text. We compare tracks and isochrones, and show how the results depend on the total CNO abundance. We notice that different initial CNO abundances between two clusters, otherwise similar in metallicity and age, may lead to differences in the turnoff morphology that can be easily attributed to an age difference. We simulate the main sequence and subgiant branch data for NGC 1851 and show that an increase of C+N+O by a factor $sim$3 best reproduces the shift between the subgiant branches. We compare the main sequence width in the color m$_{F336W}$-m$_{F814W}$ with models, and find that the maximum helium abundance compatible with the data is Y$simeq$0.29. We consider the result in the framework of the formation of the second stellar generation in globular clusters, for the bulk of which we estimate a helium abundance of Y$simlt 0.26$.
