Multiplex sampling is an extension of standard Monte Carlo methods for estimating characteristics of the distribution of a response when the response is a function of several variables, each of which comes from a known distribution. The extension is required when each variable is available in a variety of distributions. Depending on the number of different distributions for each variable and on the number of variables there are many possible different combinations each of which, in general, will give a different distribution to the response. If characteristics of the response are to be estimated for many or all of these combinations there will be a plethora of Monte Carlos to be performed if usual procedures are followed. This in turn can require a great number of observations of the response; if these are difficult to obtain, e.g., if they must be determined experimentally, the carrying out of such a program can easily prove impracticable. Multiplex sampling is a method for estimating the characteristics for all the different distributions for the response by using a relatively small number of observations. This is accomplished by sampling from an efficient sample space and then using weighted sampling formulas. The functional form for the probability density function describing this sample space is derived in a companion paper [7]; here we assume this form and consider the practical aspects.