Nawat, the highly endangered indigenous language of El Salvador, is undergoing a revitalization process. This dissertation, conceived within this context, focuses on the second-language (L2) acquisition of features of Nawat pronunciation by learners who have Salvadoran Spanish as their first language (L1). Specifically, I assess the acquisition of one Nawat segmental feature and one prosodic feature through their production by L2 learners: 1) the Nawat velar stop phoneme /g/, which has different allophones according to its surrounding sounds, and 2) Nawat utterance-final vowel glottalization, that is, the production of utterance-final vowels with increased constriction of the vocal folds to signal the end of an utterance. The goals of this dissertation are: 1) to explore whether L2 Nawat learners become more proficient in the production of these Nawat features as their years of Nawat study increase, 2) to identify the linguistic contexts in which the productions of these Nawat features by L2 learners differ the most from L1 Nawat speakers, and 3) to assess whether the progress in the L2 acquisition of these Nawat features is equally linear or if one aspect is more difficult to learn than the other. For this purpose, a total of 21 L2 Nawat learners were recorded performing a reading task in Nawat and Spanish designed to elicit the sounds of interest in different contexts. A control group consisting of five L1 Nawat speakers were also recorded performing open-ended interviews and a picture description task to serve as a baseline. I find that, overall, L2 Nawat learners become more proficient in the production of Nawat /g/ as their years of study increase. However, there are contexts in which productions of Nawat /g/ by L2 learners and L1 speakers consistently diverge, namely, when word-initial /g/ is in the post-lateral position and when word-medial /g/ is in the post-approximant, post-lateral, and post-obstruent contexts. I argue that these differences are the result of L1 transfer and hypercorrection. As for the L2 acquisition of Nawat utterance-final vowel glottalization, progress is observed but, unlike Nawat /g/, even the most advanced learners do not produce Nawat utterance-final vowels at rates comparable to L1 speakers. In fact, most Nawat utterance-final vowels produced by L2 learners show weakened voicing rather than glottalization, which I interpret as L1 transfer. Thus, an asymmetry in the acquisition of these features of Nawat pronunciation is identified, which I attribute to the inherent complexity of prosody, the perceived similarity between the L1 and L2 sound systems, frequency, and the Nawat input received by L2 learners. This dissertation highlights the need to inform the Nawat teaching curricula to focus on the production and perception of utterance-final vowel glottalization and develop more effective pedagogical materials and practices to teach them. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]