1 Introduction

In spite of the tremendous improvement of the new generation telescopes (VLT, HST, KECK etc.), severe physical limits do not allow to resolve distant stellar systems, and integrated colours and spectra will remain for a long time the only way to investigate the evolutionary history of stellar systems beyond the Local Group of galaxies. In this context, theoretical predictions concerning simple stellar populations (SSP), i.e. stellar populations with a common age and chemical composition, appear of paramount relevance, since SSP represent the fundamental ingredient to approach the observational evidences from more complex stellar systems.

In the recent past several authors have presented SSP models (see, e.g., Buzzoni 1989; Leitherer et al. 1996; Tantalo et al. 1996 (TCBF); Vazdekis et al. 1996 (VCPB); Maraston 1998; Kurth et al. 1999 (KFF)) for a wide range of ages and metallicities, already providing valuable scenarios on this matter. However, only Covino et al. (1994) make a brief discussion on the effect of different assumptions on the mass loss rate, whereas one knows that the HB morphology is critically affected by the amount of mass lost during the RGB phase (see, e.g., Iben & Renzini 1983; Jørgensen & Thejll 1993). Moreover, we still lack a detailed evaluation of the influence of statistical fluctuations on the cluster integrated colours, a problem which has been addressed only by Tantalo et al. (1996).

In a previous paper (Brocato et al. 1999a, Paper I) we have recently presented theoretical integrated colours of simple stellar populations with ages smaller or of the order of 5 Gyr, testing the results on a suitable sample of Large Magellanic Cloud stellar clusters. As discussed in that paper, the given upper limit to cluster ages was suggested by the opportunity to minimize uncertainties about mass loss which, in old stellar clusters, governs the temperature of horizontal branch (HB) stars, affecting the predicted colours. This paper will now cross this limit, exploring the sensitivity of the theoretical scenario for old stellar clusters to reasonable assumptions about the efficiency of mass loss.

Following a slightly unconventional approach, we will start our investigation by testing first the capability of the adopted evolutionary models to reproduce the CM diagram distribution of stars in selected globular clusters. This appears to us as a preliminary but relevant step, aiming to ground the investigation on reasonably firm basis. Such a test will be discussed in Sect. 3, after presenting in Sect. 2 preliminary information on the adopted evolutionary scenario and on the computational procedure. In Sect. 4 we will make use of integrated colours for a template population to investigate the effects of statistical fluctuations. According to the procedure already envisaged in Paper I, we will correlate the size of predicted fluctuations to the absolute integrated V magnitude of the clusters, allowing a quantitative estimate of the not negligible uncertainty affecting a large portion of galactic globular clusters. We will finally present theoretically predicted colours as a function of age, metallicity and/or efficiency of mass loss, making suitable comparisons with observations and with previous available results. A short discussion will close the paper.