76 abstracts: technology 



with no potential difference applied in order to check up the effect of 

 the electric stress. 



The alternating contact with the coating of air and water was obtained 

 by filling the cone and allowing the water to evaporate which took about 

 a week. Readings of the current flow were made at suitable intervals. 

 The first appearance of current flow was taken as indicating the end of 

 the useful life of the specimen as an insulating coating. 



The average life of the paints was about one hundred and sixteen days, 

 the maximum life obtained from any specimen being but little more than 

 a year. No conclusive evidence was obtained that the low potential 

 differences applied had any effect in hastening the initial failure of the 

 coatings. The wrappings lasted longer than the paints and dips but 

 none of them much more than four hundred days. It seems from the 

 results that the failure of the coatings must be caused by absorption by 

 them of water which in time penetrates to the iron, allowing current to 

 flow and destroy the coating by electrolysis. After the first appearance 

 of current flow the destruction of the coating was observed to proceed 

 very rapidly. 



The conclusion drawn from the results of the laboratory tests, i.e., 

 that the protection against electrolysis which is obtained by wrappmg 

 or painting pipes or other metallic bodies for use underground is only 

 temporary, is borne out by tests on insulated pipes buried in the ground 

 under practical conditions, and also by correspondence with gas and 

 water companies whose experiences lead to the same conclusion. 



B. McC. 



TECHNOLOGY — Industrial gas calorimetry. C. W. Waidner and E. 

 F. Mueller. To appear as a Technologic Paper of the Bureau of 

 Standards. 



After reference to the more important types of calorimeters that have 

 been applied to the measurement of the heating values of gases, the 

 principle of the flow calorimeter is set forth, and an example is given 

 showing the reductions of, and corrections to the observed data that are 

 required in order to find the total and net heating values. 



The results of an experimental investigation of laboratory gas meters, 

 showing the errors to which such meters are liable, the precautions to be 

 observed and the accuracy attainable in their use, are given at length. It 

 is shown that an accuracy of about 0.2 per cent may be attained if the 

 meter is calibrated in situ at the time of use, and that the calibration 

 may be reproduced from time to time by making suitable adjustments, 

 to an accuracy of about 0.5 per cent. 



