VOLVOCINE ALGAE have a recent evolution of multicellularity, only 30-70 million years ago. This may produce a better record of the early history of this process than we have for other multicellular organisms. Metazoans and multicellular plants evolved over 550 million years ago (multicellular plants multiple times, and some suggest a very ancient history of multicellular algae over 800 million years ago). The fossil record for fungi is not very good, but unambiguous multicellular fungi were present by 500 million years ago. Bacteria meanwhile beat everyone out by evolving multicellularity several times perhaps 2-3 billion years ago. Since most multicellular organisms have a distant origin, extinction has eroded the base of their evolutionary trees so that the details of the transition are hard to extract. The volvocine algae have a much more recent history of multicellularity, and we have been able to determine much about their evolutionary history from phylogenetic studies of these algae and their relatives.

While I tend to look at plants as more stuff for animals to eat than as interesting in their own right, the volvocine algae are strange enough to attract my attention. They’re descended from phytoflagellates, which originally caused a taxonomic quandary because in the Linnean system they were not easily categorized as either plant or animal, and ended up chucked into the miscellaneous bin of unicellular whatsits—kingdom Protista. While they carry out photosynthesis like plants, they also motor about like little submarines, detect light with the use of eyespots, and some also have heterotrophic capabilities (sometimes ingesting organic matter or bacteria). Volvocines retained some of these traits, and the video below shows the characteristic tumbling motion of a Volvox (“fierce roller” from Latin), caused by the beating of the flagella on the somatic cells of the colony wall. This motility allows volvocine algae to control their position in the water column, approach light to utilize photosynthesis, and respond to other attractive or repellant stimuli.

The transition from single-celled phytoflagellate to multicellular volvocine is often approximated as going through ancestral states similar to existing volvocines of intermediate complexity.1 The volvocines evolved from single-celled phytoflagellates similar to Chlamydomonas reinharditii (A below). The most simple volvocine algae are simple clumps of four cells, each capable of reproduction. In colonies with 8-32 cells colony morphology can be a sheet (Gonium pectorale, B), flattened sphere, or sphere. Larger colonies are uniformly hollow spheres (Eudorina elegans (C), Pleodorina californica (D), Volvox carteri (E), V. aureus (F)), and the largest colonies can contain up to 50,000 cells with cells differentiated into a large number of small somatic cells and a small number of much larger germ cells (seen clearly in E and F below).2 The germ cells asexually produce offspring colonies that are initially contained within the parent colony’s sphere, and are freed when the shell of somatic cells undergoes programmed cell death and releases the daughter colonies.

Various volvicide algae and relatives

While it is likely broadly true that volvocine evolution went through stages similar to intermediate extant species, the actual history was probably more complicated. A new study attempts phylogenetic analysis to determine the order of acquisition of traits and detect possible reversals and preadaptations.3

Various transitions found in the volvocine algae include:

  1. Incomplete cytokinesis, joining daughter cells via cytoplasmic bridges.
  2. Incomplete inversion of the embryo coordinated via cytoplasmic bridges, and placing flagella on the convex side of the sheet of cells.
  3. Rotation of basal bodies so cells are oriented in one direction.
  4. Establishment of organismal polarity.
  5. Transformation of cell walls into an extracellular matrix.
  6. Genetic control of the number of cells in a colony (achieved by limiting the number (n) of divisions, with the final number of cells equaling 2n).
  7. Complete inversion, required for spheroidal colonies to place flagella on the exterior surface of the sphere.
  8. Increased volume of the extracellular matrix.
  9. Partial germ/soma division of labor (dedicated soma cells).
  10. Complete germ/soma division of labor (dedicated soma and germ cells).
  11. Asymmetric cell division producing large and small daughter cells, promoting differentiation.
  12. Bifurcation of the cell division program to determine soma or germ fate early in development.

According to the phylogenetic tree obtained (note the rampant paraphyly and polyphyly due to an over-reliance upon particular traits in naming the algae!), the only ancestral trait to all are the extracellular matrix and the genetic control of cell number by the limitation of the number of cell divisions. Traits acquired once and never lost appear to be incomplete cytokinesis, rotation of basal bodies, and establishment of organismal polarity. It appears partial and complete inversion evolved independently. Increased extracellular matrix volume evolved once, was lost in one lineage, and was later reacquired. Some algae that do not have differentiation of soma cells appear to be descended from ancestors with partial differentiation of soma, having secondarily lost this trait.

Possible preadaptations include multiple fission found in the unicellular relatives of the volvocine algae. Incomplete cytokinesis would then produce multiple daughter cells fused together, and their attachment and close relatedness would then promote selection for cooperative behavior. This is plausible because occasionally unicellular Chlamydomonas fail to separate completely after undergoing multiple fission.1 Additionally, the aptly named Gonium dispersum appears to be able to live colonially or unicellularly, since among a clonal line after some divisions cells may swim away on their own, while in most cases the daughters remain together as a typical Gonium colony.1 Another possible preadaptation is the Chlamydomonas tripartite cell wall, with a gelatinous central layer. This layer is expanded in unicellular Vitreochlamys, and may provide an analogy for the evolution of the extracellular matrix.

Kirk originally proposed that traits 1-5 must have evolved close together in time since removal of any one of these traits in living volvocines produces a dysfunctional colony.1 However, the phylogenetic analysis disproves this hypothesis, showing, for instance, that inversion is not uniformly required, and that partial and complete inversion arose independently. The common ancestor of Goniaceae (which experiences partial inversion) and Volvocaceae (which experiences complete inversion) probably was similar to Astrephomene, lacking inversion. This study indicates again that we cannot base arguments of “irreducible complexity” upon modern systems since examination often reveals ancestral systems that lack supposedly key components.

Generally speaking, arguments of this form—that a set of traits must have evolved simultaneously because modern taxa do not function well if one is disrupted—should always be viewed with caution. Phylogenies that are missing taxa, either because of extinction or incomplete sampling, will often appear to show that multiple traits arose simultaneously. This effect is demonstrated by the example of the Tetrabaenaceae: any phylogenetic reconstruction missing these two species will appear to show that ECM [extracellular matrix] and genetic modulation of cell number arose at the same time as incomplete cytokinesis, basal body rotation, and establishment of organismal polarity. Another way of showing this is to remove taxa from a known phylogeny. For example, if Gonium is omitted, expanded volume of ECM will appear to have arisen simultaneously with the establishment of organismal polarity in the MRCA [most recent common ancestor] of Goniaceae + Volvocaceae (Figs. 3A, E). We might then wrongly conclude that a spherical colony with cells arranged on the periphery is required for the establishment of organismal polarity. If enough taxa are removed, any number of traits will appear to have arisen simultaneously. Character state changes do not occur in isolation, and numerous unmeasured characters likely change to accommodate each change in a measured character. We should not, therefore, be surprised when disruption of a trait that arose some millions of years ago disrupts the functioning of the organism, nor should we conclude from this that the trait arose simultaneously with other changes that have occurred in the same lineage.

The recruitment of some cells to soma required loss of reproductive capability for these cells, and is only favored when reproductive cheating can be controlled. Cheating was probably initially suppressed by control of the number of cell divisions, which is ancestral for these algae. This would prevent out-of-control replication of cheater cells. Control of cell number must have been a prerequisite for expansion of the extracellular matrix, which is energetically costly to produce, and provides an opportunity for freeloaders to utilize neighboring cells’ matrix while not contributing their own resources to it. Following the control of cell number incomplete cytokinesis, rotation of basal bodies, and the establishment of polarity integrated individual cells into a cooperative colony. In some partial or complete inversion developed.

However, evolution of soma may have required additional selection. This appears to be favored in larger colonies, either due to a benefit of division or labor or a trade-off between mobility and fecundity. Cells that are reproducing lose their flagella, which is detrimental for colony motility. Interestingly, partial differentiation (dedicated soma) was lost in two Eudorina species with small colony size (n=5). Dedicating cells to soma may be detrimental for reproduction in small colonies. Complete differentiation is only found in two sister species, V. obversus and V. carteri, which while large (n=11) are not among the largest colonies (n=15). It is possible that algae with partial differentiation are undergoing a second cycle of selection and must further suppress somatic cheats before complete differentiation can develop.

The major hurdle in evolving multicellularity appears to be the control of cheaters. The modern volvocine algae are in different stages of evolving multicellularity, with some being simple colonial aggregates of cells, and others being truly multicellular organisms with dedicated germ and soma lineages. The volvocine algae may be in the process of completely integrating their individual cells into a multicellular union, and more study of the mechanisms of cheater suppression in these algae should provide more insight into the requirements for the evolution of multicellularity.

The evolution of multicellularity was also covered recently at Sandwalk and Pharyngula. The evolution of multicellularity also shares some properties (such as suppression of cheaters) with the evolution of social colonies in insects such as ants, which I discussed recently.


  1. Kirk, D. L. “A twelve-step program for evolving multicellularity and a division of labor.” BioEssays 2005, 27, 299-310. DOI:10.1002/bies.20197
  2. Michod, R. E. “Evolution of individuality during the transition from unicellular to multicellular life.” Proceedings of the National Academy of Sciences 2007, 104, 8613-8618. DOI:10.1073/pnas.0701489104
  3. Herron, M. D.; Michod, R. E. “Evolution of complexity in the volvocine algae: transitions in individuality through Darwin’s eye.” Evolution 2008, 62, 436-451. DOI:10.1111/j.1558-5646.2007.00304.x
Advertisements