In research performed by the authors thousands of biaxial stress measurements on samples of stressed canvas and discarded paintings were collected with the help of a specially built automated logging instrument. A project is described in which a particular canvas sample was exposed to controlled fluctuations in temperature and RH, and the resulting changes in tension were registered. The sample was then wetted, sized, primed, and again exposed to similar fluctuations in temperature and RH following each treatment. Plotted graphically, these measurements give a clear record of the modifications in stress created by each of the above treatments. A sample of similar canvas primed about 100 years ago was also tested in the same way. The test results led to the conclusion that the resistance of the substrate to deformation (Modulus E) was of the utmost importance for the preservation of the paint film it supported. Accordingly, this Resistance to Deformation (E) of stretched canvas was measured within the range of 5-100% RH and 0-50° and at various canvas tensions. While small changes of the Modulus E due to environmental fluctuations were detected, tension was clearly the decisive factor in the resistance of the canvas to deformation. In pursuing this finding it was determined that canvas was incapable of maintaining its tension for any length of time, particularly when exposed to high RH. However, at low tension, canvas was incapable of withstanding the expansions and contractions of a stiff paint film. The detrimental effect of temperature fluctuations, heretofore underestimated, was noted in these measurements. In addition, these tests documented the beneficial effect of the self-adjusting, constant tension stretcher. This stretcher maintains sufficient tension within the canvas by absorbing stress peaks and preventing overstretching during stress fluctuations.