Objective: Several genome-wide screens have been performed in autism spectrum disorders resulting in the identification of numerous putative susceptibility loci. Analyses of pooled primary data should result in an increased sample size and the different study samples have a potential to strengthen the evidence for some earlier identified loci, reveal novel loci, and even to provide information of the general significance of the locus. The objective of this study was to search for potential susceptibility loci for autism, which are supported by two independent samples. Methods: We performed a combined analysis of the primary genome scan data of the Autism Genetic Resource Exchange (AGRE)and Finnish autism samples to reveal susceptibility loci potentially shared by these study samples. Results: In the initial combined data analysis, the best loci (p < 0.05) were observed at 1p12- q25, 3p24- 26, 4q21- 31, 5p15- q12, 6q14- 21, 7q33- 36, 8q22- 24, 17p12q21, and 19p13- q13. The combined analysis of Finnish and AGRE families showed the most promising shared locus on 3p24- 26 with nonparametric logarithm of odds (NPL) score of 2.20 (p = 0.011). The combined data analysis did not provide increased linkage evidence for the earlier identified loci on 3q25- 27 or 17p12- q21. However, the 17pl2- q21 locus remained promising also in the combined sample (NPLall = 2.38, p = 0.0076). Interpretation: Our study of 314 autism families highlights the importance of further analyses on 3p24- 26 locus involving comprehensive molecular genetic analyses of oxytocin receptor gene (OXTR), a positional and functional candidate gene for autism. Objective: Several genome-wide screens have been performed in autism spectrum disorders resulting in the identification of numerous putative susceptibility loci. Analyzes of pooled primary data should result in an increased sample size and the different study samples have a potential to strengthen the evidence for some had identified loci, reveal novel loci, and even to provide information of the general significance of the locus. The objective of this study was to search for potential susceptibility loci for autism, which are supported by two independent samples. Methods: We performed a combined analysis of the primary genome scan data of the Autism Genetic Resource Exchange (AGRE) and Finnish autism samples to reveal susceptibility loci potentially shared by these study samples. Results: In the initial combined data analysis, the best loci (p <0.05) were observed at 1p12- q25, 3p24-26, 4q21-31, 5p15- q12, 6q14-21, 7q33-36, 8q22-24, 17p12q21, and 19p13- q13. The combined anal ysis of Finnish and AGRE families showed the most promising shared locus on 3p24-26 with nonparametric logarithm of odds (NPL) score of 2.20 (p = 0.011). The combined data analysis did not provide increased linkage evidence for the earlier identified loci on 3q25 - 27 or 17p12- q21. However, the 17pl2- q21 locus remained promising also in the combined sample (NPLall = 2.38, p = 0.0076). Interpretation: Our study of 314 autism families highlights the importance of further analyzes on 3p24-26 locus involving comprehensive molecular genetic analyzes of oxytocin receptor gene (OXTR), a positional and functional candidate gene for autism.