『Abstract
　Based on ten heavy metals collected twice annually at 59 sites from 1998 to 2004, enrichment factors (EFs), principal component analysis (PCA) and multivariate linear regression of absolute principal component scores (MLR-APCS) were used in identification and source apportionment of the anthropogenic heavy metals in marine sediment. EFs with Fe as a normalizer and local background as reference values was properly tested and suitable in Hong Kong, and Zn, Ni, Pb, Cu, Cd, Hg and Cr mainly originated from anthropogenic sources, while Al, Mn and Fe were derived from rocks weathering. Rotated PCA and GIS mapping further identified two types of anthropogenic sources and their impacted regions: (1) electronic industrial pollution, riparian runoff and vehicle exhaust impacted the entire Victoria Harbour, inner Tolo Harbour, Eastern Buffer, inner Deep Bay and Cheung Chau; and (2) discharges from textile factories and paint, influenced Tsuen Wan Bay and Kwun Tong typhoon shelter and Rambler Channel. In addition, MLR-APCS was successfully introduced to quantitatively determine the source contributions with uncertainties almost less than 8％: the first anthropogenic sources were responsible for 50.0, 45.1, 86.6, 78.9 and 87.5％ of the Zn, Pb, Cu, Cd and Hg, respectively, whereas 49.9％ of the Ni and 58.4％ of the Cr came from the second anthropogenic sources.

Keywords: Heavy metals; Marine sediment; Source apportionment; Enrichment factors; Multivariate statistics; GIS』

Introduction
Materials and methods
　Study area and sampling methods
　Enrichment factors
　Multivariate analyses and data transformation
Results and discussion
　Data pretreatment
　Differentiation between natural and anthropogenic elements
　Potential source identification
　Source contributions for each element
Conclusion
Acknowledgments
References