Academics, having written their peer reviewed articles, may at some stage in the make their work Open Access (OA). They can do this by self-archiving an electronic version of their article to a personal or departmental web page or to an institutional or subject repository, such that the article then becomes freely available to anyone with Internet access to read and cite. Those authors who do not wish to do this may leave their article solely in the hands of a toll access (TA) journal publisher who charges for access, consigning their article to remain behind a subscription barrier. Lawrence (2003), in a short study, noted that conference articles in computer science that were freely available on the World Wide Web were more highly cited that those that were not. Following this, there have been a number of studies which have tried to establish whether peer-reviewed articles from a range of disciplines which are freely available on the World Wide Web, and hence are OA, accrue more citations than those articles which remain behind subscription barriers (Antelman 2004, Davis and Fromerth 2007, Eysenbach 2006, Harnad and Brody 2004, Kurtz and Henneken 2007, Moed 2007). These authors generally agree that there is a citation advantage to those articles that have been made OA, but are either uncertain about, or find that they cannot agree on, the cause of this advantage. The causes of this citation advantage could simply be that OA articles are available well in advance of formal publication, and so have a longer period in which to accrue citations, or simply that more authors, because they are freely available, can read and cite them. As part of this debate, Smith (2007) asked whether authors from developing countries might contribute to higher citation counts by accessing OA articles and citing them more readily than TA articles. As part of a larger study of the citation advantage of OA articles (Norris, Oppenheim and Rowland 2008), research was undertaken to see whether a higher proportion of citations to OA articles came from authors based in countries where funds for the purchase of journals are very limited. Mathematics was chosen as the field to be studied, because no special programme for access in developing countries, such as HINARI (2007), covers this subject. The results show that the majority of citations were given by Americans to Americans, but the admittedly small number of citations from authors in developing countries do seem to show a higher proportion of citations given to OA articles than is the case for citations from developed countries. Some of the evidence for this conclusion is, however, mixed, with some of the data pointing toward a more complex picture of citation behaviour.