This section is from the book "Cyclopedia Of Architecture, Carpentry, And Building", by James C. et al. Also available from Amazon: Cyclopedia Of Architecture, Carpentry And Building.
There is a very general prejudice against the use of the ordinary Portland cement for setting limestone, sandstone, and marble, the theory being that the moisture from the mixture, in getting away, travels through the stone, carrying with it certain staining matter, which is either deposited on the surface or unites chemically with ingredients in the air to stain the face. Thus there have come to be used certain materials of much less structural value than the Portland cement. These are called non-staining cements, and are sold at a high price. There are also on the market many materials intended to be put on the back of stone, which the manufacturers claim will prevent the passage of moisture and staining matter. While it can easily be demonstrated that water containing cement running over the face of soft stones does leave distinct traces, it is a different thing to show that such stone or marble can be stained by materials in the joints or on the back. It is well, before any amount of money is spent, to experiment in producing the stain on the stone used with the ordinary cement.
A small box can be filled with soft mortar, and a piece of stone or marble laid therein so that the top surface is 1/8, inch above the mortar. Then, by tacking a. piece of waterproof paper to the box sides, and cutting a hole somewhat smaller than the upper face of the stone or marble sample used, and by weights bringing the inner edge of the paper to the stone so that practically no moisture can pass except through the sample and off from the exposed surface, the results of the so-called staining qualities of ordinary cement can be observed.