Conventional wisdom asserts that the primary benefit of multicellularity, hence presumably what drove its evolution, is the division of labor, or specialization, provided by differentiated cells. (see Maynard Smith, J. & Szathmáry, E. The Major Transitions in Evolution, 1995). But differentiation arose slowly. Prior to the emergence of differentiation there was a more temporary form of specialization in biofilms; there was polymorphic messaging between genetically identical cells (e.g., quorum sensing); there was apoptotic behavior; and there was stigmergy. So it is difficult to argue what the "primary" benefit might have been.
The transition from single-cell to multicellular life did not happen in one evolutionary leap. We do not know with much precision when and how the strategies arose that support today's multicellular organisms nor what alternatives were tried and failed. All we know is that what we see today survived the test of time.
From our current perspective it appears that multicellular life evolved from single cells in two stages. First, single cell organisms evolved the ability to form loose cooperative communities, called biofilms, that can perhaps be thought of as “training wheels” for multicellular life. Perhaps the earliest colony bacteria were the cyanobacteria that evolved more than three billion years ago. Their fossil remains are visible today because these colonies secreted a thick gel as protection from solar radiation unattenuated by (then nonexistent) atmospheric ozone. This gel, in turn, trapped sand and debris from the surf which, together with lime secreted by the bacteria, formed the beautiful patterns of the Stromatolite fossil reefs visible in Australia (see image at left). These structures vary in size from twig-size to semi-truck size. Biofilms remain common today. Present-day examples of biofilms include slime mold, dental plaque, films on rocks in streams and many more.
The need for four key principles of multicellularity
Perhaps one billion years ago true multicellular organisms formed – plants, animals, and fungi – known generically as Metazoans. All cells in a Metazoan organism share the same DNA. As the organism develops, the cells specialize by sequestering and permanently silencing much of their DNA according to developmental genetic programs. Some organisms have multiple stages of stable forms, e.g., insects that exhibit larval, pupae, and adult forms. But these developmental stages all involve programmed cell differentiation. For most cells (stem cells being the exception) differentiation and the resulting cell specialization, is dramatic and irreversible.
While the full complement of genes and DNA control sequences in the multi-cellular genome is far more complex than that of single cell organisms, any given type of cell – and there are about 250 different types in humans – is functionally much simpler. That is, specialization and differentiation during the evolution of multicellularity lowers the complexity of individual cells. Each differentiated cell type expresses (i.e., makes the proteins coded for) only a few percent of the 25,000 total human genes. For example, all cells in the body have the gene for hemoglobin, but only red blood cells express it. Along with this specialization, the cells had to send messages to each other without passing DNA. They also had to cooperate to develop their "body," which is a stigmergy structure. And they had to develop apoptosis to remove cells that had outlived their usefulness or become dangerous. Without all four of those principles, true multicellularity would not work.
Might a different set of basic multicellular principles have worked just as well? Possibly. A few small differences in the early stages might have given rise to other principles. But, if so, we cannot know what they might have been.