As neurons extend their axons, it is thought that newly synthesised membrane components travel in vesicles along the axon, fuse with the growth cone membrane, and diffuse back along the axonal membrane. However, it is difficult to explain how axons continue to be populated with membrane proteins as they extend in length. To investigate this problem, we have used a CEPU-green fluorescent protein (GFP) chimeric protein to study the site of insertion of new glycosyl phosphatidyl inositol (GPI)-anchored glycoproteins and their subsequent behaviour in chick dorsal root ganglia (DRG) neurons. Infection of cultures grown for 24 h revealed rapid expression of CEPU-GFP over the whole surface of the neuron, more rapidly than could be accounted for by diffusion from the growth cone, and fluorescence intensity was uniform along the length of the neurite. Photobleaching experiments of neurite membrane revealed that recovery of fluorescence was due to diffusion from adjacent membranes and there was no evidence for membrane flow in either direction. Photobleaching of membrane adjacent to the cell body also showed rapid recovery, with chimera diffusing both from cell body membrane and the distal neurite membrane into the bleached area. These results suggest there is no barrier to diffusion between the cell body and neurite membrane in DRG and sympathetic neurons cultured for 1 or 2 days in vitro. We propose that the neurite is populated by newly synthesised chimera by diffusion from both regions. This situation may also occur in neurons in the early stages of extending axons in vivo prior to polarisation and the development of the dendritic field.