Water fleas on cycles

Article metrics

Abundances of animals vary over time, and explaining these fluctuations is a central goal of ecology. Especially intriguing are the many cases where populations cycle1. The period of cycles is rarely one year, ruling out obvious tracking of the seasons. This leaves at least two possibilities to explore. Predators may over-exploit their prey, leading to cycles of population explosion and collapse, with prey numbers rising to high levels during the latter phase. Alternatively, partially synchronized reproduction might drive demographic cycles with alternating large and small cohorts of offspring, leading to cycles in adult numbers. On page 653 of this issue2, McCauley et al. tease apart these possibilities for an organism with well-known population cycles — the common water flea Daphnia.

In thinking about population cycles, ecologists often turn to mathematics3. In mathematical models, predator–prey cycles generally have high amplitudes4, which are greatly exaggerated by enriching the prey's habitat5. In contrast, theory predicts that cycles driven by demography and phased reproduction lack the high amplitude and sensitivity to enrichment characteristic of predator–prey cycles.

In field studies, it is difficult to disentangle the mechanisms driving population cycles from the many other confounding sources of natural variation, and experimental control is often infeasible. Model systems of predator and prey cultured in the laboratory have been popular — such as Daphnia and the microalgal prey that it filters from the water. The properties of this system would be expected to give classical predator–prey cycles. Daphnia populations can double within weeks and over-exploit their prey, clearing natural waters of algae6. Algae reproduce even more rapidly, doubling within days, and easily reach high abundance when Daphnia are sparse. But decades of study of Daphnia and algae have not shown clear evidence for predator–prey cycles.

Early laboratory studies of Daphnia populations did indeed reveal cycles7, but regular additions of algal food prevented over-exploitation. Instead, Daphnia reproduction became synchronized. When sparse, adults obtained a large ration of food and produced numerous offspring. Adults of the subsequent generation were abundant, obtained a much smaller ration, and left fewer offspring. Cycles with relatively low amplitude, and out-of-phase variations of juveniles and adults, were the result. Extending these studies, McCauley and his colleagues2 have conducted experiments and combed the literature on natural populations of Daphnia. Their work supports the presence of demographic, but not predator–prey cycles. Why is the basic biology of this predator and its prey, which strongly hints of over-exploitation and prey escape, so misleading? Previous painstaking work by this group has eliminated several hypotheses8.

One remaining hypothesis postulates that algal species relatively invulnerable to ingestion by Daphnia have a selective advantage in rich habitats. Inedible algae compete for nutrients with edible algae, and effectively reduce the richness of the habitat for edible algae, dampening the tendency to cycle. McCauley and colleagues achieved the difficult task of preventing inedible algae from proliferating in their laboratory systems. Doing so induced classical predator–prey cycles, and this result does much to settle the question of why Daphnia do not show such cycles under less controlled conditions. This finding probably generalizes to many predator–prey systems: invulnerable competitors of prey should be highly favoured by natural selection, and therefore common in nature.

Perhaps more intriguing, when inedible algae were absent, only some of the experimental cultures displayed predator–prey cycles of high amplitude. Others cycled with lower amplitude, and with phasing of adults and juveniles suggesting demographic cycles. This result demonstrates ‘alternative attractors’ — that is, two dynamical states that characterize the long-term behaviour of Daphnia populations, with subtle differences in initial conditions determining which state is reached. The nonlinear mathematical models used to study population cycles often predict such phenomena, but most ecologists have probably despaired of verifying their existence in real populations.

McCauley and colleagues also reveal a biological mechanism governing the switch between the two types of population cycles. Daphnia are parthenogenetic — they reproduce without sex — and their normal eggs quickly hatch into juveniles. Daphnia also produce, by sexual reproduction, resting eggs for long-term survival (Fig. 1). This diverts energy from normal reproduction and could prevent population explosions and over-exploitation, thus turning off the predator–prey cycle. McCauley and colleagues tested this possibility by replacing mothers carrying resting eggs one-for-one with mothers carrying normal eggs. This nearly doubled the amplitude of population cycles.

Figure 1: Reproduction in Daphnia.

Normal eggs are produced parthenogenetically and incubated in the dorsal brood chamber (top). Resting eggs are provisioned with energy stores and enclosed in a protective brood case (bottom). McCauley et al.2 found that these resting eggs have the effect of damping the species' population cycles.

Production of resting eggs essentially disperses offspring to the future, and is analogous to dispersing them in space. In theory, spatial dispersal could stabilize or destabilize population dynamics, and the range of possibilities for temporal dispersal requires further study. Long-term population dynamics could depend on the hatching of resting eggs. Indeed, others have drawn attention to similarities between the ‘egg banks’ of aquatic animals and the ‘seed banks’ of plants9. Dispersal to the future is a widespread feature among animals, in the form of resting eggs, diapausing stages (in which animals go into a kind of suspended animation), and long-lived adults. It could stretch out the timescale of ecological dynamics, blurring the distinction with evolutionary dynamics. The study by McCauley et al. should do much to stimulate research on its implications in many types of organisms.


  1. 1

    Kendall, B. E., Prendergast, J. & Bjørnstad, O. N. Ecol. Lett. 1, 160–164 (1998).

  2. 2

    McCauley, E., Nisbet, R. M., Murdoch, W. W., de Roos, A. M. & Gurney, W. S. C. Nature 402, 653–656 (1999).

  3. 3

    Volterra, V. Nature 118, 558–560 (1926).

  4. 4

    May, R. M. Science 177, 900–902 (1972).

  5. 5

    Rosenzweig, M. L. Science 171, 385–387 (1971).

  6. 6

    Lampert, W., Fleckner, W., Rai, H. & Taylor, B. E. Limnol. Oceanogr. 31, 478–490 (1986).

  7. 7

    Slobodkin, L. B. Ecol. Monogr. 24, 69–88 (1954).

  8. 8

    Murdoch, W. W. et al. Ecology 79, 1339–1356 (1998).

  9. 9

    Hairston, N. G. Jr Limnol. Oceanogr. 41, 1087–1092 (1996).

Download references

Author information

Correspondence to James P. Grover.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Grover, J. Water fleas on cycles. Nature 402, 592–593 (1999) doi:10.1038/45104

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.