The Brunauer-Emmett-Teller (BET) isotherm equation is probably the most widely used analytic equation in the range 0.01 less than or similar to P/P-0 less than or similar to 0.3 where it provides a standard measure of microscopic surface area. Here P is the pressure above the adsorbed layer, and P-0 is the vapor pressure of the adsorbate, both at the temperature of the measurement. Once two measurable constants have been obtained the BET equation permits the calculation of the entire isotherm from Henry’s law at low pressures and coverages to the vapor pressure at multilayer coverages. A critical comparison between the entire equation and experimental data requires data over a very broad range of pressure and coverage. These were essentially unavailable before the advent of ultrahigh vacuum and, even today, are rare. But experimental data published many years ago by the author for physical adsorption isotherms of argon, krypton, and xenon on an adsorbent of porous silver at T = 77.4 K are suitable for a test of the whole BET equation that was not carried out at the time. The present article fulfills that omission. It is found that, while the BET equation is an adequate description of the data in the range 0.01 less than or similar to P/P-0 less than or similar to 1, serious and increasing divergencies occur as the pressure decreases further. A review of ultrahigh vacuum isotherm data for helium and nitrogen leads to the same conclusion for these gases also. The special case of hydrogen is important in predicting the base pressures in modern colliders. Here, however, the vapor pressure of hydrogen at 4.2 K is about 3.2x10(-7) Torr, and magnitude some four orders in pressure below this (required for a decisive test of the BET equation) is already well into the range where quantitative hydrogen measurements are difficult. Hence conclusions about the applicability of the BET equation to hydrogen isotherm data are less cer;ain. An analytic equation based, in part, on the Dubinin-Radushkevich isotherm at low pressures, which was proposed earlier, gives a good description of the argon, krypton, and xenon data on porous silver over the entire measured range, some thirteen orders of magnitude in pressure.