This thesis is a report of voice onset time (VOT) of word-initial ejectives, implosives, and plain stops in Mopan, Itzaj, and Yukateko Maya and could be considered a preliminary study of VOT in the three languages. Although many studies on VOT and/or ejectives and plain stops, and some on implosives and VOT were found, most of them were on non-Mayan language families like Athabaskan (Gordon, 1996; Gordon, Potter, Dawson, De Reuse, & Ladefoged, 2001; Lindau, 1982; McDonough & Ladefoged, 1993; McDonough & Wood, 2008; Wright, Hargus, & Davis, 2002), Caucasian (Maddieson, Rajabov, & Sonnenschein, 1996), Indo-Aryan (Nihalani, 1991), Kwa and Chadic languages spoken in Nigeria and the surrounding area (Lindau, 1980), Sino-Tibetan (Zeng, 2008), or a mixture of non-Mayan languages from several families (Ladefoged & Cho, 2001). Phonetic/phonological studies on Yukatekan-Mayan languages (Academia de Lenguas Mayas de Guatemala, 2001; Archibald, 1996; Blair, 1964; Hofling, 2000; McQuown, 1967; Orie & Bricker, 2000; Robertson, 1977; Schumann-Gálvez, 1997, 2000) did not focus on VOT or ejectives, implosives, and plain stops.

In this study, a total 314 utterances of bilabial implosives, bilabial, alveolar, and velar ejectives, and bilabial, alveolar, and velar plain stops in word-initial position from one Itzaj, five Yukateko, and four Mopan speakers were measured from the stop release to the onset of voicing of the following vowel, and results were compared to VOT of other languages with similar phonemes. All three languages have fairly long average (negative) VOTs for implosives, contradicting Schumann-Galvez's claim that both Itzaj (2000) and Mopan (1997) have /b/ rather than /b'/. No voiceless implosives were found in any of the three languages. Of the three languages, Mopan had the longest average VOT for all phonemes and Yukateko had the shortest. Itzaj had the greatest difference between VOT of plain stops and those of ejectives and implosives and Mopan had the smallest, so VOT is a "perceptual cue" to aid in differentiation of phonemes to varying extents (Gordon et al., 2001, p. 424). Compared to other languages, Mopan and Itzaj had the longest VOT for velar ejectives, surpassing even the Athabaskan languages which have a reputation for long ejective and aspirate VOTs. Two unexpected results were that Yukateko had a very long VOT for plain velar stops, which was only 6 ms shorter than its velar ejective and that the average VOT of the plain bilabial voiceless stop in Mopan was actually longer than the corresponding ejective. Further study to determine the cause of these anomalies is warranted.