Just why the public has not been educated into a correct understanding of the gas situation is hard to say, unless it be that an inbred prejudice against believing the word of any corporation has led to an utter repudiation of such explanatory statements as may emanate from time to time from the gas office. And it must be admitted that many of the explanations are misleading, either through the intention of the superior officials or by reason of the ignorance of their subordinates. Hardly has the chill of shortening days driven us indoors in the early twilight before complaints of poor gas become epidemic. Now, what _is_ ‘poor’ gas? Is the gas deficient in light-giving constituents, or is it merely burned in such a manner as not to afford a satisfactory illumination? The charter of Greater New York requires that the illuminating gas supplied throughout the city shall be of at least twenty candle power, or illuminating quality, or richness--that is to say, if we burn this gas in a standard burner at the standard pressure (or at as near this pressure as may be), so that the rate of consumption is five cubic feet an hour, the flame thus produced shall be equivalent to twenty standard sperm candles, each burning at the rate of one hundred and twenty grains of sperm per hour, and all bunched--if such a thing were possible. There can be hardly any doubt but that all the gas sent out from modern gas works fulfills the above requirement. Indeed, my own tests give results ranging from twenty-two to twenty-eight candles, with an average of about twenty-four. Manifestly, the gas _sent out_ is not ‘poor.’ Nevertheless, the fact that the gas as manufactured is of the required candle power is no indication that the product as delivered to the consumer will give a similarly satisfactory test. Distribution of gas is attended with many perplexities, not the least of which is condensation. The illuminating hydrocarbons, or light-giving constituents held in suspension in the gas, are not so firmly fixed therein as to be unaffected by the size of the pipe, the character of the internal pipe surface, and barometric and thermometric variations. The transmission of gas causes, therefore, a loss of candle power ranging from a small fraction to several candles, although it is possible to conceive of conditions so extraordinarily favorable that the illuminating quality of the gas might be actually improved by distribution.