This dissertation contributes new data to the debate on the source and manner of transfer in third language (L3) acquisition. The "L1 Status Factor" (L1SF, e.g., Hermas, 2010; Leung, 2002; Lozano-Pozo, 2003) maintains that the transfer source is the native language (L1), whereas the "L2 Status Factor" (L2SF, Bardel & Falk, 2007) maintains it is the second language (L2). The "Typological Primacy Model" (TPM, e.g., Giancaspro, Halloran, & Iverson, 2015; Rothman, 2010, 2011) proposes that the source of transfer is determined by (perceived) structural/typological similarity between the L3 and either the L1 or the L2; it also proposes that the transfer happens in a wholesale manner at the initial stages of L3 acquisition. Others propose property-by-property transfer, as in the "Linguistic Proximity Model" (e.g., Westergaard, 2021; Westergaard, Mitrofanova, Mykhaylyk, & Rodina, 2017) and the "Scalpel Model" (Slabakova, 2017). These hypotheses are tested in this study with less-studied L3 learner populations: L1Cantonese-L2English-L3Mandarin (CEM, n = 32) learners and L1Korean-L2English-L3Mandarin (KEM, n = 34) learners. Moreover, the dissertation also explores whether L2 proficiency--a potentially important yet understudied factor--plays a role in determining the source of transfer. The learners completed--in addition to English and Mandarin proficiency tests and a background questionnaire--acceptability judgment tasks (AJTs) in both their L2 (English) and their L3 (Mandarin). Native speakers of the four languages involved serve as controls, completing the AJT in their respective L1 as well as a background questionnaire. There are five linguistic phenomena tested: (a) the Double Object Construction (DOC, e.g., "John gave Mary a letter"); (b) the Prepositional Dative Construction (PDC, e.g., "John gave a letter to Mary"); (c) the Reverse PDC (e.g., *"John gave to Mary a letter"); (d) Passivization of the Recipient (POR) in the DOC (e.g., "Mary was given a letter by John"); and (e) Passivization of the Theme (POT) in the DOC (e.g., *"A letter was given Mary by John"). Crucially, the learners' L1s (i.e., Cantonese or Korean) and their L2 Target Language (English) behave differently across these constructions. There are three principal findings in this study: First, results from both the CEM learners and the KEM learners suggest that the source of transfer is the L1 (with one exception on a single phenomenon by a KEM learner). The transfer patterns that emerged in the CEM and KEM learners of this study are therefore compatible with both the L1SF and the TPM (with, unfortunately, no way to decide between the two). Second, since the L1 is overwhelmingly implicated as the transfer language, it is thus unsurprising that no significant correlation was found between L2 proficiency and the source of transfer. Third, again as a consequence of transfer being, in essence, restricted to the L1, no evidence was found to support the property by property transfer hypothesis, and hence the transfer patterns of the CEM and KEM learners are compatible with the wholesale transfer hypothesis. [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.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]