Most biological membranes are extremely complex structures consisting of hu
ndreds or even thousands of different lipid and protein molecules. The prev
ailing view regarding the organisation of these membranes is based on the f
luid-mosaic model proposed by Singer and Nicholson in 1972. According-to th
is model, phospholipids together with some other lipids form a fluid bilaye
r in which these lipids are diffusing very rapidly laterally. The idea of r
apid lateral diffusion implies that, in general, the different lipid specie
s would be randomly distributed in the plain of the membrane. However, ther
e are recent data indicating that the components tend to adopt regular (sup
erlattice-like) distributions in fluid, mixed bilayers. Based on this, a su
perlattice model of membranes has been proposed. This superlattice model is
intriguing because it allows only a limited certain number of "critical" c
ompositions. These critical compositions could play a key role in the regul
ation of the lipid compositions of biological membranes. Furthermore, such
putative critical compositions could explain how compositionally distinct o
rganelles can exist despite of rapid inter-organelle membrane traffic. In t
his review, these intriguing predictions are discussed along with the basic
principles of the model and the evidence supporting it. (C) 1999 Elsevier
Science B.V. All rights reserved.