440 H. A. Hazen — Condensing Hygrometer 



preciable. It is easy to see this, suppose the temperature of 

 the liquid be lowered 'l per second, and there is a large retard- 

 ation in the diminution of the outside temperature, or, which is 

 the same thing, the outside temperature is higher because of 

 the air being 40° above the liquid ; in such case the thermome- 

 ter, we will say, will have a temperature -1° lower than the 

 dew-point, or in other words it takes 1 second for the tempera- 

 ture of the liquid to reach the outside of the plate. Now if we 

 lower the temperature at the rate of '05° per second then the 

 inside temperature will be -05° lower than the outside approx- 

 imately. If now we consider the dew-point constant during 

 the experiment, we shall have with the rapid fall an indicated 

 dew-point '05° lower than with the other. A long series of 

 trials have shown the difference in the two methods inappre- 

 ciable. It should be borne in mind that whatever be the 

 amount of the effect due to the thickness of the plate, the ten- 

 dency will always be toward a too low dew-point. A method 

 of testing this question which would seem to be free from many 

 of the objections that may be raised against those already pro- 

 posed is the following, lack of time and funds have prevented 

 carrying out the plan. We may arrange a plate with two or 

 even three different thicknesses of metal. Suppose such a 

 plate have thicknesses for each third of its width of '01", '02", 

 ■04" and it be cooled clown gradually ; we would expect dew 

 first on the thinnest and last on the thickest, and the difference 

 in temperature as the dew appeared on each could be readily 

 noted. 



A second source of error may be due to a difference in the 

 hvgrometric properties of plates. It is readily seen that if there 

 be such a difference it would be rather a difficult matter to set- 

 tle which should be taken as the standard, though it would 

 seem that, so far as this one thing is considered, that plate 

 which gave the lowest dew-point would be the best. No ex- 

 periments have yet been made to settle this point. It would 

 require observations, from two trained men, at the same time. 

 It may be said, however, that this consideration is far out- 

 weighed by others relating to the cleaning of the plate, 

 detection of dew, etc. It is a very difficult matter to clean the 

 plate properl} 7 ; the best method would seem to be by using a 

 mixture of dilute ammonia and rouge as a wash, applying with 

 silk, thoroughly drying and wiping it off by rubbing; this 

 process gives a most beautiful surface for the dew. Quickness 

 of detection of dew can only be obtained by practice. No 

 fixed rules can be laid down for arranging the light, the posi- 

 tion of the eye, etc. It is necessary to cut off all reflections 

 from light-colored objects ; this may best be done by placing a 

 black cloth in front of the apparatus. A light breeze is an ad- 



