So, apparently all sorts of cells can produce "ultra-weak" photons, in species ranging from plants to human beings. References in the Fels article indicate that this has been known since the 1980's, which leaves me feeling somewhat indignant that I never learned about biophotons in any of my classes. I'm going to quote the paper because it has a nice summary of previous biphoton studies (click the link to the article if you'd like to see all of the citations -- PLoS ONE is open access, so I know everyone can see the Fels paper, but reading the other references might be difficult if you don't have institutional subscriptions to their sources):
Although biophotons may carry biologically relevant information , , , only very little is known about whether individuals indeed use them for sending and receiving information. A few studies (with populations separated from each other molecularly but not electromagnetically) strongly suggest biophotons as transmitters of information: e.g., onion roots influence mitosis positively in neighbouring onion roots (supposedly due to so-called mitogenetic radiation , being probably effective in the UV-range ); yeast cells, which emit biophotons in the UV- and the visible range , affect growth in other yeast cells positively ; tissue cells arrange themselves in a non-random manner according to the pattern of tissue cells on the opposite side of a glass slide ; and germinating Fucus-zygotes probably sense biophotons emitted by their living substrate to which they direct their growth .
Thus, it seems that many simple cells are about to "see" photons emitted by their neighbors, and respond to these electromagnetic signals through changes in their growth patterns.
With this background, Fels decided to test the effect of putative biophotonic communication in the single-celled organism Paramecium caudatum. The experiment was simple: he cultured the Paramecia in glass and quartz containers called cuvettes -- an outer cuvette containing one culture, with a smaller cuvette placed inside it that contained another culture. The different cuvette materials allowed different wavelengths of light to pass between the cultures. He placed the cuvettes into a dark box to control for the effect of other light sources in the environment. After giving the Paramecia some time to grow, he then observed the effects of one culture on its neighbor, to see if the organisms were able to transmit information between cuvettes even though no molecular signals were able to diffuse across the barriers. Cell growth and cell division (measured by counting the total number of Paramecia in each culture at the end of the experiment) and feeding (measured by vacuole formation) were quantified, to see if electromagnetic signals between cultures could influence these simple behaviors. The study found that:
[P]opulation growth and the feeding rate of Paramecium caudatum depended significantly on (i) the presence or absence of a neighbouring population, (ii) the number of cells in the neighbouring population and (iii) the material (glass or quartz) separating these populations. The results strongly support the existence of a non-molecular information-carrying system that is based on photons.
In other words, the culture adjacent to the one being measured has an effect on the growth and feeding behavior of a group of Paramecia, and this effect is dependent on the material that separates the two cultures. Specifically, when analyzing growth, Fels saw that "large populations grew significantly better (than controls) when separated with glass from the small neighbour populations, but they grew as well as the controls when separated with quartz from the smaller neighbour populations." For feeding behavior, he reports: "When separated by quartz from a few neighbouring cells (15–20), vacuole formation was higher than for the glass units, but when separated from many neighbours (300–400 cells) it was the lowest of all treatments. The opposite effect was found for populations separated by glass." This suggests that the effects on growth were largely explained by feeding behavior -- the populations that exhibited increased growth also seemed to have a higher rate of feeding.
Are we sure that this effect is due to biophoton transmission? Not exactly. However, the differences observed between populations separated by glass and those separated by quartz would seem to indicate that at least some of the effect can be explained by the different wavelengths transmitted through those materials (glass serves as a filter for some wavelengths of UV light; quartz allows them to pass). This seems to rule out other possible mechanisms for information transfer, such as a gaseous molecular signal that diffused into the air around the cultures, or an infrared (heat) effect, as it's not likely that these would be dependent on the filtering effects of the separating material.
Interestingly, Fels reports a difference in cell growth based not only on the material separating the two cultures, but on the outer material. Glass and quartz were used to form both the inner and outer cuvettes, yet only the inner cuvette should have had an effect on electromagnetic transmission between the two cultures (the outer cuvette's job was just to keep Paramecia from spilling everywhere). The article claims that the nature of this effect is outside the scope of this study, which is a reasonable position to take, but if the cultures are interacting with the material that houses them in a way that doesn't involve biophotonic communication, this presents a potential confound for the results.
Of course, under natural conditions, single-celled organisms or cells within a single organism are not separated from each other by glass/quartz, and use a variety of molecular signals for communication. Even if biophotons play a role in cellular communication, their effects may be modulated by molecular signals (or vice versa). It would be interesting to figure out the mechanism by which biophotons are generated or received and to selectively disrupt it in Paramecia, to see if that had an effect on growth and feeding behavior between populations separated by glass/quartz or by a more permeable barrier. This would allow us to further dissect the effects of electromagnetic signaling in the presence or absence of molecular signaling, and to better understand what those crazy biophotons are really doing.
Fels, D. (2009). Cellular Communication through Light PLoS ONE, 4 (4) DOI: 10.1371/journal.pone.0005086