Hydrothermal vent organisms are adapted to their extreme and patchily distributed habitats. They are expected to have evolved mechanisms that keep them in their specific habitation. Since little is known about the recruitment or habitat selection of HV organisms such as brachyurans, we examined the properties of several hydrothermal vent-associated cues on the behavior of the hydrothermal vent (HV) crab Xenograpsus testudinatus in the laboratory that were contrasted by the offering of non-vent cues. This crab species is endemic and dominates the vent fauna of Turtle Island off the NE coast of Taiwan. HV crabs were separately and in combination offered the following vent-specific cues: (1) sulfuric sediment, (3) air-bubbling, (4) elevated temperature, (5) dead settled zooplankton, (7) other crabs, and (8) shade. The non-vent-specific cues were: (2) quarz sediment, (6) dead fish, (8) light. These cues were provided on either side of a two-choice chamber. The movement of individual crabs was monitored: as initial and final choices, and as the proportion of time the crabs spent in each compartment (resident time). Cues were offered alone and no such cue as a control in the same set-up. Sulfuric sediments and dead fish were significantly more attractive to females, and other crabs irrespective of gender were significantly more attractive to males. When compared to expected distributions, crabs, irrespective of gender, significantly avoided light and tended to select other crabs, air-bubbling, sulfuric sediment, elevated temperature, dead fish, dead zooplankton, and quarz sediments in the order of decreasing importance. Data do not support the hypothesis that dead settled zooplankton was particularly attractive nor that the other gender was selected. A combination of several vent-associated cues (sulfuric sediment, elevated temperature, air-bubbling) facilitated the strongest attraction to the crabs as reflected by all response variables. The 'first choice' responses were always consistent with the side of the choice compartment in which they spent the longest amount of time (resident time), but not with the 'final choice' of crabs, suggesting that the 'resident time' in addition to their 'first choice' is more reliable than just the 'final choice'. The present results provide the first indication that several vent-associated habitat cues function as attractors for HV crabs. Habitat choice is also reflected by crab larval distribution in the field which tend to stay near the bottom not to be carried away from their specific habitat.