The subject of this study is the pre-main sequence (PMS) evolution of stars less massive than ~1.2 solar mass, using lithium as the main tool for confronting model predictions with observations. In addition, it is proposed to use Li as a test of the possible substellar nature of brown dwarf candidates. The main results of this work are: 1.) The initial Li abundance in a sample of 48 weak-line T Tauri stars is log N(Li)NLTE=3.11 +/- 0.05 2.) No PMS Li depletion is observed in stars with masses 1.2-1.0 solar mass. 3.) In the mass range 0.9-0.7 solar mass, there is large PMS Li depletion, but it seems to be inhibited in the presence of rapid rotation 4.) In stars with masses ~0.4 solar mass, the Li abundances fall abruptly at luminosities below 0.5 L_odot and rotation does not seem to play any role 5.) No published PMS evolutionary model quantitatively accounts for the Li abundances observed in our stellar samples 6.) Below the substelar mass limit (0.08 solar mass) Li is less efficiently destroyed. Our computations show that at masses /= 0. 8.) We have obtained intermediate resolution spectra of 23 very low-mass stars or candidate brown dwarfs with the main aim of applying the Li test. In none of the observed stars have we obtained a positive detection, implying strong depletion of Li in them. We infer lower limits to their masses, typically M >/= 0.063 solar mass. This results constrain the substellar domain in the H-R diagram and the evolutionary models at the end of the main sequence. (SECTION: Dissertation Summaries)