The study of resolved stellar populations provides a powerful tool to follow galaxy evolution directly in terms of physical parameters such as age (star formation history, SFH), chemical composition and enrichment history, initial mass function, environment, and dynamical history of the system. Photometry of individual stars in at least two filters and the interpretation of Colour-Magnitude Diagram (CMD) morphology gives the least ambiguous and most accurate information about variations in star formation rate (sfr) within a galaxy back to the oldest stars (see Figure 1). Some of the physical parameters that affect a CMD are strongly correlated, such as metallicity and age, since successive generations of star formation may be progressively enriched in the heavier elements. Careful, detailed CMD analysis is a proven, uniquely powerful approach (e.g., Tosi et al. 1991; Tolstoy & Saha 1996; Aparicio et al. 1996; Mighell 1997; Dohm-Palmer et al. 1997a, b; Hurley-Keller et al. 1998; Gallagher et al. 1998; Tolstoy et al. 1998; Tolstoy 1998) that benefits enormously from the high spatial resolution, high quality imaging. HST has led the way in recent years, and we are optimistic that ground based telescopes capable of achieving excellent seeing, such as VLT, will also provide much useful data, especially in the blue, and in relatively sparse stellar systems such as dwarf irregular (dI) galaxies.
The small dI and dSph galaxies are considered the most likely connection to higher redshift, late type evolving systems, which makes them the most interesting to look at when trying to make a connection to the high redshift universe. In the LG they appear to exhibit a wide variety of SFHs. These results have affected our understanding of galaxy formation and evolution by demonstrating the importance of episodic star formation in nearby low mass galaxies. The larger galaxies in the LG have evidence of sizeable old halos, which appear to represent the majority of star formation in the LG by mass, although the problems distinguishing between effects of age and metallicity in a CMD result in a degree of uncertainty in the exact age distribution in these halos. High quality, deep imaging is the only way of extending detailed SFH studies past the Galaxy and its satellites, enabling us to obtain a picture of the fossil record of star formation in galaxies of various types and sizes, and to identify both commonalities and differences in their SFH across the LG to be able to make the comparison with cosmological surveys (Tolstoy 1998) more accurate.
The Local Group (LG) will provide a picture of the global star formation properties of galaxies with a wide variety of mass, metallicity, gas content etc., and will make a sample that ought to reflect the SFH of the Universe which can be compared to high redshift survys (e.g., Madau et al. 1998). Initial comparisons suggest these different approaches do not yield the same results (Tolstoy 1998; Fukugita et al. 1998), but the errors are large due to the lack of detailed SFHs of nearby galaxies. Careful determinations of the complete SFH of nearby galaxies is thus extremely important for understanding star formation properties of nearby galaxies in detail and also to make the connection to high-redshift studies. It is still not clear if bursting star formation is common enough in LG dIs to account for a large population of compact emission line galaxies at intermediate redshifts. If the conclusions of Madau et al. and the SFH of the LG still do not agree then it will be of critical importance for the field of high redshift galaxy research that this discrepancy is understood. It might be that redshift surveys are strongly biased towards low mass dI galaxies undergoing bursts of star formation, and thus they are not accurate indicators of the dominant mode (by mass) of star formation in the Universe.