Time-lapse imaging of cells has been historically limited to a few hours because of the difficulty associated with maintaining the appropriate physiological pH and temperature of the cells during prolonged imaging. Carbon dioxide produced by cellular respiration changes the pH of the medium, and sodium bicarbonate based buffers typically used for cell culture require an atmosphere of 5% CO2 to maintain a physiologically appropriate pH. Microscope stage incubators can provide such an atmosphere in an encapsulated box set up on the microscope stage, but researchers often complain about temperature instability and gradients that arise in such systems. Additionally, the price and complexity of such systems led us to develop an inexpensive alternative consisting of a miniaturized microscope and stage that fits inside a conventional CO2 incubator. In addition to the optical and mechanical parts, I created a userfriendly computer program that allows for the design of custom imaging sessions and protocols, developed autofocusing algorithms, and the associated electronics for instrument control and communication between the computer and the microscopes. I placed special emphasis on giving the microscopes the capability to perform fluorescent imaging because fluorescence has become the workhorse of molecular studies in cell biology in recent times. My primary use of the microscopes that I built centered on studies of a human-derived mammary cancer cell line, the MCF cell progression model of cancer, which provides a model of increasing tumorigenicity in cells all arising from the same genetic background. By tracking single cells I find that cancer cells have heterogeneous subpopulations of cells with varied division potential, mitotic errors and aberrant nuclei, which I quantified and categorized. I describe a method for quantifying the absence of contact inhibition of growth in cancer cells in cocultures, and I find that cancer cells affect the morphology of normal cells. I find that the cancer cells are better at creating invasive strands in Matrigel, and these invasions are temporally correlated with the change of cell culture medium. Additionally, an absence of serum causes the retraction of normal cells but promotes the invasion of the cancer cells.