The sharpnose pufferfish Canthigaster valentini is widespread over the Indo-Pacific tropical coral-reef region. The social system contains territorial male, non-territorial male, and territorial female. Territorial male always occupy a fixed territory and own several territorial females forming a harem in its territory. Previous researches were concerned about field observations or laboratory tests for agonistic respondences towards self-mirror stimuli. Research on dyad interaction is rarely seen in literature. The first objective of this study focuses on the agonistic interactions between two individuals from different combinations of the abovementioned (i.e., male-male and female-female) social categories under a more spacious laboratory condition. Characteristics for intra-individual behavioral sequences were described and compared. Besides, attentions were also paid to the possible factors (e.g., sex, social status) controlling the major aggressive display (i.e., Circling). The second objective is to conduct research on the water spitting behavior of this species that I had discovered. Factors including body size, presence of food and depth of water were tested to see if any of them is the releasing factor for this behavior. The amount of information transmitted in an intra-individual behavior sequence for a male winner was highest, whereas that for a female winner was lowest. Except for Circling, FRD, FLD, and HD were the other major aggressive behaviors. The average number of aggressive behavior in an intra-individual behavior sequence and display time for male were higher than that for female. More attacks were launched in female pairs than in male pairs. For female pairs, attacks were performed by both contestants , whereas only the winners attacked its opponent in male pairs. Male tended to launch the first attack after the longest Circling but that situation did not hold for female. Circling , a agonistic behavior nor quantitatively described in other studies, can be classified into S-type and C-type. The probability of occurrences for these two types of circling were not significant different. Cycle number was higher in the early phase of an aggressive interaction in male pairs. And male spent more time in circling during an aggressive interaction. Under the experimental setting of this study, body length did not affect the outcomes of a aggressive contest. For male pairs, ratio of body high/body length played a role in determining the outcomes. Presence of food , water depth, and movement of objects above the holding aquarium were all releasing factors for the water spitting behavior and presence of food might be the most important one.