This dissertation offers new approaches to an old and well-known problem in the study of the development of Romance varieties: duplicate lexis or doublets. Traditional analyses of duplication are narrow in scope both in what qualifies as a doublet (the popular/learned opposition has dominated, to the exclusion of other pairs) and in channels of transmission, particularly mixing as distinguished from, and an alternative to, "borrowing." I apply theories of diffusion and sociolinguistics to the production of duplicate lexis, which tends to develop historically along a continuum between synonymous variants, or "Doppelformen," and semasiolexically polarized "Scheideformen," with virtually no semantic overlap. Chapter one examines the period before c.1200, beginning with the pan-Romance duplicative reflexes of PENSARE and RATIONEM, and other doublets based in sound changes germane to Castile (e.g., "popar" [similar to] "palpar" ) arguing that these forms were the result of incomplete diffusion, mixing, and ultimately reallocation, rather than a Latinizing phonolexical stratum. Chapter two treats the era of between c.1200-1500, in which multiple input varieties of Peninsular Romance and Medieval Latin enrich lexicons with new phonolexical variants. In several doublets, however, "Latinism" phonologically coincides with said input varieties (e.g., "plegar" [similar to] "llegar"). Chapter three examines duplication between the 16th and 18th centuries, a period of increasing standardization practices, focusing on doublets that result from re Latinization as defined by Harris-Northall (1999, 2006) (e.g., "afeccion" [similar to] "aficion") and the combined results of both koineization and standardization in the Americas (e.g., "fierro" [similar to] "hierro"). Thus examining the lexical development of Ibero-Romance in Castile and elsewhere, we gain new panchronic insights into the nature of doublet production as related to dialect contact, mixing and semasiolexical reallocation of complex coetymological lexis, from "Doppelformen" to "Scheideformen." This study advances an approach to phonolexical duplication anchored in Wright's (1982, 2010) model of Latin/Romance, theories of diffusion (Wang 1969) and sociolinguistic theories of both koineization in medieval Spanish (Tuten 2003) and early standardization of Spanish (Harris-Northall 1996, 1999; Tejedo-Herrero 2008), providing a much needed panchronically-oriented analysis of doublet production wherein doublets not easily explained as dialect-borrowing or Latinism are illuminated in terms of their development. [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.]