Student Theses and Dissertations

Date of Award


Document Type


RU Laboratory

Muir Laboratory


claustral ecology, population dynamics, closed ecosystems, Escherichia coli, Chlamydomonas reinhardtii, Tetrahymena thermophila


For almost any species in any environment, it is nearly impossible to predict its fitness from molecular knowledge. If fitness is not to be a mere tautology, reproducible measurements of the survival and reproduction of populations are needed over many generations. Laboratory microbial ecosystems afford the short time and length scales required for such measurements. Their conventional implementations, batch cultures with period refreshment of growth medium or chemostats with continuous refreshment, have a number of disadvantages, such as the introduction of additional frequencies, selection for surface growth and the distortion of chemical interactions. In closed ecosystems free energy is instead supplied as light, allowing for simpler, replicable protocols and a consistent interpretation of interactions, independent of their mode or timescale. Here, I describe a model closed ecosystem consisting of three singlecelled microbes, Escherichia coli, Chlamydomonas reinhardtii and Tetrahymena thermophila and show that these species can coexist for hundreds of days under closure. Using a custom built in situ fluorescence microscopy set up, the densities of these three species can be measured automatically and noninvasively over months with low classification error and large dynamical range. When kept under identical boundary conditions, these ecosystems reproducibly diverge in composition, with characteristic divergence times of ~20 days for T. thermophila, ~40 days for the other two species, and an approximately linear increase of an aggregate divergence measure over the first ~60 days. For two ecosystems, densities were measured continuously under constant conditions and their dynamics shown to be nonstationary for all three species >100 days after closure. As a consequence, conventional time series methods assuming stationarity are inadequate and wavelet analysis is proposed as an alternative. Species-species interactions are further investigated using oscillations in illumination intensity. Densities of C. reinhardtii and, surprisingly, E. coli respond to modest perturbations of light intensity. Variation of the modulation frequency strongly implicates the circadian clock of C. reinhardtii in its response. The nonlinearity of the E. coli response suggests that it depends on C. reinhardtii density or spatial distribution rather than directly responds to the modulation of illumination. Further improvements in the detection of interactions are proposed.


A thesis presented to the faculty of The Rockefeller University in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

Permanent URL

Included in

Life Sciences Commons