It is evident that in the 4 billion years of evolution no prokaryote evolved a cellular complexity that even vaguely resembles that of a eukaryote, other than the archaeal cell that acquired the mitochondrion through endosymbiosis. Thoughts on the origin of the endomembrane system are linked to thoughts on the origin of eukaryotes themselves. Current models traditionally derive the origin of the endomembrane system from inward invaginations of the plasma membrane, such that the endoplasmic reticulum (ER) lumen is topologically homologous to the environment. But prokaryotic vesicle flux is outward not inward and such models also fail in explaining the transition from an archaeal plasma membrane based on isoprene ethers to a bacterial-type membrane based on fatty acid esters. We propose that the endomembrane system stems from outer membrane vesicles (OMVs) the mitochondrial endosymbiont secreted. In terms of the number and nature of evolutionary innovations required to evolve a basic endomembrane system our minimal premises can hardly be underbid, and account for: (i) the transitional mechanism that converted the composition of eukaryotic membranes from archaeal to bacterial lipids, (ii) the finding that eukaryotic lipid synthesis occurs predominantly at the ER and mitochondria, (iii) the circumstance that eukaryotes store Ca2+ in the ER, (iv) the formation of the nucleus from the ER, not vice versa, (v) the archaeal ancestry, localisation, and orientation of the eukaryotic V-ATPase in food vacuoles and other things we will discuss. From our proposal, a natural evolutionary order in the origin of several key characters of eukaryotic cells unfolds in that, during eukaryogenesis, the ER represented the first autogenous (nonendosymbiotic) cell compartment, formed from OMVs secreted by the mitochondrion. Endosymbiotic theory directly accounts for the origin of eukaryote complexity and it required a ‘garden variety’ archaeal host to do so, as some theories for eukaryote origin predicted.