The patterns of chemical abundances, as derived in the atmosphere of the stars 
or in the interstellar medium of galaxies, keep the imprint of the various 
physical processes which take place and influence the interstellar medium of 
galaxies during their evolution. In the first part of the talk, I will present 
how we model the chemical evolution of some classical and ultra-faint dwarf 
spheroidal galaxies of the Milky Way, starting from the chemical abundances of 
alpha and n-capture elements which are observed in their member stars at the 
present time. Thanks to the very efficient spectrographs currently working in 
connection with large and medium-sized telescopes, which have provided us large 
amounts of accurate data, the stellar populations of these galaxies have been 
studied in always greater detail in the last years. In the second part of the 
talk, I will focus on the (N/O) vs. (O/H) abundance pattern, as inferred in a 
sample of SDSS galaxies (Data Release 7, Abazajian et al. 2009). This dataset - 
integrated with the N and O abundances derived in metal-poor, star forming 
dwarf galaxies - spans a wide metallicity range, enabling us to recover the 
trend of the (N/O) vs. (O/H) relation with a definition never reached before. 
This collection of data clearly demonstrates the existence of a plateau in the 
(N/O) ratios at very low metallicity, followed by an increase which steepens as 
the metallicity increases. This trend has been interpreted in the past as the 
signature of two complementary producers of nitrogen acting in galaxies (pure 
primary N from massive stars + primary and secondary N from low- and 
intermediate-mass stars). I will present a set of novel chemical evolution 
models to reproduce such trend.