Learners are becoming increasingly divers. They may have much personal, social, cultural, psychological, and cognitive diversity. Forming suitable learning groups represents, therefore, a hard and time-consuming task. In Mobile Computer Supported Collaborative Learning (MCSCL) environments, this task is more difficult. Instructors need to consider many more issues, such as the rapid change of mobile learners' context, their direct and natural interaction, and the characteristics of mobile devices and networks. This paper presents a systematic literature review (SLR) that examines the relevant solutions for the problem of group formation in MCSCL environments. In the context of this SLR, an initial list of 178 papers was reviewed. After careful analysis of each paper using specific selection criteria and a quality assessment method, a final list of 12 relevant studies was filtered and used to answer the research questions. The findings revealed that: (a) there is a lack of approaches addressing the group formation problem in MCSCL environments; (b) the most proposed solutions do not allow instructors to customize the grouping process; (c) there is no useful solutions to automatically capture and evaluate many of learners' behaviours and context information; (d) the majority of approaches do not support a dynamic formation of learning groups; (e) the majority of approaches do not provide descriptions about the implemented grouping algorithms nor about the evaluation methods. Extracted and synthesized data from the selected studies is discussed in this paper, together with current research gaps and recommendations for further works.