The molecules and cells of the immune system tend to fall into two categories — classical and non-classical. This month's issue contains several articles that focus on the non-classical category. For example, as described by Branch Moody and Steven Porcelli on page 11, CD1 molecules are MHC class-I-like molecules that present antigens to T cells — but, the antigens are glycolipids rather than peptides, the antigen-binding groove is different and CD1 molecules traffic through the cell using two separate, but parallel, pathways. The authors speculate that the endosomal pathway presents foreign glycolipids from exogenous sources preferentially, whereas self-lipids are processed mainly in non-endosomal compartments.

Macrophages are described classically as being activated by TH1 cytokines, but, as Siamon Gordon explains on page 23, this categorization is somewhat limited, because macrophages can also be activated in an 'alternative' manner by the TH2 cytokines IL-4 and IL-13.

Mice are the classical animal model of choice for many researchers, but large animal models have contributed to immunological research also, as outlined by Wayne Hein and Philip Griebel on page 79.

Two articles in this issue focus on antibodies. On page 63, Sidonia Fagarasan and Tasuku Honjo discuss the induction of IgA responses in the gut and the physiological role of IgA, which is the most common antibody type to be produced at mucosal surfaces. The autoantibody repertoire is surprisingly restricted, and it is not entirely clear why this is the case. On page 73, Paul Plotz suggests that the selection of autoantigens is governed by their ability to interact with 'gateway' receptors on antigen-presenting cells.