Believe it or not, this appears to have something to do with their circadian rhythms!
Back in the 1960s and early 1970s, there was quite a lot of research published on the circadian rhythms in earthworms, mostly by Miriam Bennett. As far as I can tell, nobody’s followed up on that work since. I know, from a trusted source, that earthworms will not run in running-wheels, believe it or not! The wheels were modified to contain a groove down the middle (so that the worm can go only in one direction and not off the wheel), the groove was covered with filter paper (to prevent the worm from escaping the groove) and the paper was kept moist with some kind of automated sprinkler system. Still, the earthworms pretty much stood still and the experiments were abandoned.
Dr.Bennett measured locomotion rhythms in other ways, as well as rhythms of oxygen consumption, light-avoidance behavior, etc. With one of my students, some years ago, I tried to use earthworms as well – we placed groups of worms in different lighting conditions (they were inside some soil, but not deep enough for them to completely avoid light) – the data were messy and inconclusive, except that worms kept in constant light all laid egg-cases and all died (evolutionary trade-off between longevity and fecundity, or just a last-ditch effort at reproduction before imminent death?). Worms in (short-day and long-day) LD cycles and in constant dark did not lay eggs and more-or-less survived a few days.
I intended to write a long post reviewing the earthworm clock literature, but that was before I got a job….perhaps one day. But the news today is that there is a new paper that suggests that clocks may have something to do with a behavior all of us have seen before: earthworms coming out to the surface during or after a rain.
In the paper, Role of diurnal rhythm of oxygen consumption in emergence from soil at night after heavy rain by earthworms, Shu-Chun Chuang and Jiun Hong Chen from the Institute of Zoology at National Taiwan University, compared responses of two different species of earthworms, one of which sufraces during rain and the other does not. They say:
Two species of earthworms were used to unravel why some earthworm species crawl out of the soil at night after heavy rain. Specimens of Amynthas gracilis, which show this behavior, were found to have poor tolerance to water immersion and a diurnal rhythm of oxygen consumption, using more oxygen at night than during the day. The other species, Pontoscolex corethrurus, survived longer under water and was never observed to crawl out of the soil after heavy rain; its oxygen consumption was not only lower than that of A. gracilis but also lacked a diurnal rhythm. Accordingly, we suggest that earthworms have at least two types of physical strategies to deal with water immersion and attendant oxygen depletion of the soil. The first is represented by A. gracilis; they crawl out of the waterlogged soil, especially at night when their oxygen consumption increases. The other strategy, shown by P. corethrurus, allows the earthworms to survive at a lower concentration of oxygen due to lower consumption; these worms can therefore remain longer in oxygen-poor conditions, and never crawl out of the soil after heavy rain.
So, one species has low oxygen consumption AND no rhythm of it. It survives fine, for a long time, when the soil is saturated with water. The other species has greater oxygen consumption and is thus more sensitive to depletion of oxygen when the ground is saturated with water. Furthermore, they also exhibit a daily rhythm of oxygen consumption – they consume more oxygen during the night than during the day. Thus, if it rains during the day, they may or may not surface, but if it rains as night they have to resurface pretty quickly.
Aydin Orstan describes the work in more detail on his blog Snail’s Tales, and he gets the hat-tip for alerting me to this paper.
Chuang, S., Chen, J.H. (2008). Role of diurnal rhythm of oxygen consumption in emergence from soil at night after heavy rain by earthworms. Invertebrate Biology, 127(1), 80-86. DOI: 10.1111/j.1744-7410.2007.00117.x