Eukaryotic cell chemotaxis, or directed cell migration in response to a chemoeffector gradient, plays a central role in many important biological process such as wound healing, cancer metastasis, and embryogenesis. In vivo, cells migrate on fibrous ECM, but chemotaxis studies are typically conducted on flat substrates which fail to recapitulate ECM or 3D gel environments with heterogeneous and poorly defined biophysical properties. To address these challenges, this thesis focused on developing a microfluidic assay device which utilizes a reductionist approach to study single cell chemotaxis on aligned, suspended ECM-mimicking nanofibers. The device is comprised of a network of microfluidic mixing channels which produce a temporally invariant, linear chemical gradient over nanofiber scaffolds in an observation channel. The microfluidic device design was guided by a numerical model and validated with experimental testing. This device was used to study mouse embryonic fibroblast NIH/3T3 response to platelet derived growth factor (PDGF) on flat polystyrene and suspended, polystyrene nanofibers with small (15 μm), and large (25 μm) spacing. Cell aspect ratio is lowest for flat polystyrene (spread morphology) and highest for large-spaced fibers (spindle morphology). Cells migrating on fibers begin to show a chemotaxis response to a PDGF gradient 10 times shallower than that required for chemotaxis response on a flat substrate. Furthermore, cells with spindle morphology maintain a robust and strong response over a broad range of chemoattractant concentration. These cells also had a 45% increase in speed and 26% increase in persistence over cells on flat polystyrene. The findings of this thesis suggest that 2D substrates may not be sufficient for studying physiologically relevant chemotaxis. / Master of Science