SCOPE AND ACCURACY These relatively simple test methods may be used in the field when seeking some positive evidence to justify expenditure for evaporation loss reduction under the following circumstances: 1. If it is not possible to carry out the preferred@ more accurate methods outlined in API Bulletin 2512. 2. If the risk element in the decision is not great. 3. If the contemplated expenditure does not warrant the cost of carrying out the procedures given in API Bulletin 2512. Although the degree of accuracy obtainable by these tests is probably not high@ the tests must be performed under comparative conditions to reduce the influence of interfering side effects as much as possible. If@ for example@ a floating blanket is being evaluated@ two similarly sized and exposed tanks containing approdimately equal levels of the same stock should be tested simultaneously ?C one equipped with the blanket and one not so equipped. If two tanks are not so available@ then one tank may be tested successively under similar conditions with and without the blanket installed. If two competitive conservation devices are being considered@ the effectiveness of each must first be established and these two evaluations then compared to determine the preferred device. It is preferable to carry out the tests on standing tanks. If this is impossible@ then the same flow conditions should prevail as closely as practical during the periods of measurement. The more nearly the test conditions correspond@ the more reliable will be the results obtained. The accuracies of the tests described herein are not positively known. The limits indicated in each case are estimates of possible reliability providing the tests are performed with proper attention to sampling and to use of instruments under the most practical comparative conditions. Nonetheless@ accuracy is an important factor in considering the results obtained. For example@ if a given device shows at 15-percent reduction in loss according to a test method having an accuracy of 20 percent@ the result must be considered inconclusive. If@ however@ this same test indicates a loss reduction of 50 percent@ then it may reasonably be concluded the device in question effects a positive improvement. The greater the number of tests performed and averaged for an evaluation@ the higher is the permissible degree of confidence in the final determination. If the result obtained by one test method is inconclusive because of its limits of accuracy@ an alternate test method may provide the positive evidence sought. Adaptability and ingenuity may be necessary to resolve the actual problem faced@ for the test methods included herein do not exhaust the possibilities for applying the comparative principle to the evaluation of loss reduction mechanisms. It is intended that they also serve as examples to suggest other similar procedures for providing data for an evaluation@ depending on the facilities available and the particular conditions which prevail. For instance@ it may be desired to evaluate an internal floating cover in a tank when no previous data has been collected and no control tank is available for a comparative test. In such a case@ application of the appropriate evaporation loss equations or charts of API Bulletin 2518: Evaporation Loss from Fixed-Roof Tanks will give the expected filling loss had the floating cover not been in use. One of the vapor analysis-type procedures herewith may been in use. One of the vapor analysis-type procedures herewith may then be employed to determine the filling loss of the tank under test equipped with the floating cover. A comparison of these two loss figures will provide a measure of efficiency of the conservation device. Although the lack of a strictly comparable control system reduces the degree of accuracy of this evaluation@ a quantitative result is obtained where otherwise no measure would be available.