The primary aim of this dissertation was to investigate the relationship between speech perception and speech production difficulties among Japanese second language (L2) learners of English, in their learning complex syllable structures. Japanese L2 learners and American English controls were tested in a categorical ABX discrimination task of nonsense words sequences (e.g., /spani/ vs /s[schwa]pani/) and included /sp, sk, pl, kl, bl, gl, spl, skl/ clusters. In the second study, production data on these same contrasts were collected by employing the delayed imitation task where speakers were asked to produce the target words in a short sentence. Productions were evaluated by American English listeners. In addition, the Versant test, a short test of English fluency by phone, was administered to see how Japanese participants' present English fluency level would correlate with their performance on the current experiments. Results of the perception experiment showed that overall accuracy by the Japanese group was significantly poorer than for Americans (Median = 71% and 100% correct, respectively). Certain clusters were harder than others for Japanese listeners. Specifically, overall accuracy in the /bl/ clusters was significantly low (Median = 63% correct). The production experiment demonstrated that, as was the case with the perception experiment, the American group showed a ceiling effect for all types of consonantal sequences. In contrast, the Japanese group's performance was consistently lower (Mean = 64% correct). Specifically, Japanese participants had difficulty producing the voiced stops + [schwa] + l tokens (e.g., /b[schwa]lani/) accurately. Interestingly, the major errors in these clusters were deletion of schwas. Correlational analyses between perception and production performance were conducted. Overall, the Japanese group's perception and production was correlated (rho = +7.82, p less than 0.01, one-tailed). Additionally, overall Versant test score was correlated with perception performance (rho = +0.470, p less than 0.01) and production performance (rho = +0.633, p less than 0.01). These results suggest that there is a link between L2 perception and L2 production at phonotactic level of acquisition. However, a picture of such link is much more complicated, as suggested by great variability among Japanese participants' performance. That is, the current study suggests that inaccuracy in L2 production is a product of interactions among inaccurate L2 perception, motor constraints of unfamiliar sequences of phonemes, and learners' individual differences in English fluency skills. [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://www.proquest.com/en-US/products/dissertations/individuals.shtml.]