The aim of this study was to compare the achievement of prospective primary science teachers in a problem-based curriculum with those in a conventional primary science teacher preparation program with regard to success in learning about gases and developing positive attitudes towards chemistry. The subjects of the study were 101 first year undergraduate students, who were in two different classes and who were taught by the same lecturer. One of the classes was randomly selected as the intervention group in which problem-based learning (PBL) was used, and the other as the control in which conventional teaching methods were used. The data were obtained through use of the gases diagnostic test (GDT), the chemistry attitude scale (CAS), and scales specific to students' evaluation of PBL such as the peer evaluation scale (PES), self evaluation scale (SES), tutor's performance evaluation scale (TPES) and students' evaluation of PBL scale (SEPBLS). Data were analysed using SPSS 10.0 (Statistical Package for Social Sciences). In order to find out the effect of the intervention (PBL) on students' learning of gases, independent sample t-tests and ANCOVA (analysis of co-variance) were used. The results obtained from the study showed that there was a statistically significant difference between the experimental and control groups in terms of students' GDT total mean scores and, their attitude towards chemistry, as well as PBL has a significant effect on the development of students' skills such as self-directed learning, cooperative learning and critical thinking.