SUMMARYDinoflagellate bioluminescence serves as a model system for examining mechanosensing by suspended motile unicellular organisms. The response latency, i.e. the delay time between the mechanical stimulus and luminescent response, provides information about the mechanotransduction and signaling process, and must be accurately known for dinoflagellate bioluminescence to be used as a flow visualization tool. This study used a novel microfluidic device to measure the response latency of a large number of individual dinoflagellates with a resolution of a few milliseconds. Suspended cells of several dinoflagellate species approximately 35 μm in diameter were directed through a 200 μm deep channel to a barrier with a 15 μm clearance impassable to the cells. Bioluminescence was stimulated when cells encountered the barrier and experienced an abrupt increase in hydrodynamic drag, and was imaged using high numerical aperture optics and a high-speed low-light video system. The average response latency for Lingulodinium polyedrum strain HJ was 15 ms (N>300 cells) at the three highest flow rates tested, with a minimum latency of 12 ms. Cells produced multiple flashes with an interval as short as 5 ms between individual flashes, suggesting that repeat stimulation involved a subset of the entire intracellular signaling pathway. The mean response latency for the dinoflagellates Pyrodinium bahamense, Alexandrium monilatum and older and newer isolates of L. polyedrum ranged from 15 to 22 ms, similar to the latencies previously determined for larger dinoflagellates with different morphologies, possibly reflecting optimization of dinoflagellate bioluminescence as a rapid anti-predation behavior. Supplementary material available online at

In this paper we discuss the design and evolution of Trinity College's ALVIN robot, an autonomous ground vehicle that has participated in the Association for Unmanned Vehicle Systems International Intelligent Ground Vehicle Competition (IGVC) since 2000. The paper first discusses the Trinity Robot Study Team, which has been responsible for developing ALVIN. We then illustrate the four generations of ALVIN, focusing on improvements made as the result of performance shortcomings and outright failures. The discussion considers the robot's body design, drive system, sensors, navigation algorithms, and vision systems. We focus especially on the vision and navigation systems developed for Trinity's fourth-generation IGVC robot, ALVIN IV. The paper concludes with a plan for future work on ALVIN and with a discussion of educational outcomes resulting from the ALVIN project.

No abstract