The generation of reporter lines for observing lens differentiation in vivo demonstrates a new strategy for embryological manipulation and allows us to address a long-standing question concerning the timing of the onset of differentiation. Xenopus tropicalis was used to make GFP reporter lines with 1-crystallin promoter elements directing GFP expression within the early lens. X. tropicalis is a close relative of X. laevis that shares the same ease of tissue manipulation with the added benefits of a diploid genome and faster life cycle. The efficiency of the Xenopus transgenic technique was improved in order to generate greater numbers of normal, adult transgenic animals and to facilitate in vivo analysis of the crystallin promoter. This transgene is transmitted through the germline, providing an accurate and consistent way to monitor lens differentiation. This line permitted us to distinguish models for how the onset of differentiation is controlled: by a process intrinsic to differentiating tissue or one dependent on external cues. This experiment would not have been feasible without the sensitivity and accuracy provided by the in vivo reporter. We find that, in specified lens ectoderm transplanted from neural tube stage donors to younger neural-plate-stage hosts, the onset of differentiation, as measured by expression of the crystallin/GFP transgene, is delayed by an average of 4.4 hours. When specified lens ectoderm is explanted into culture, the delay was an average of 16.3 hours relative to control embryos. These data suggest that the onset of differentiation in specified ectoderm can be altered by the environment and imply that this onset is normally controlled by external cues rather than by an intrinsic mechanism.