This study examined the number concentrations, size distributions, shapes, and polymer compositions of microplastic particles (MPs) in Tokyo Bay surface waters with a neuston net (350 mu m mesh) and a pump system (10 mu m mesh). The results demonstrated that the pump system collected a broader range of MPs, particularly smaller and fibrous particles, leading to higher concentrations than the neuston net. In contrast, the neuston net was less effective at capturing MPs < 1000 <mu>m, especially fibrous MPs. Principal component analysis revealed that the polymer composition differed between the two methods, with poly(ethylene terephthalate) (PET) and poly(methyl methacrylate) (PMMA) being dominant in pump system samples, while polyethylene (PE) and polypropylene (PP) were predominant in neuston net samples. These differences were influenced by multiple factors, including mesh size, sampling depth, and the volume of water filtered. The findings highlight the limitations of single-method sampling and suggest that integrating multiple sampling approaches is essential for a more comprehensive understanding of MP distribution and characteristics in surface waters. Future research should refine sampling strategies to improve the accuracy of MP pollution assessments and better inform ecological risk evaluations.