22 MR. J. H. CKINDI.KV <>\ AN EXPERIM FATAL INVESTIGATION OF 



This result appeared very interesting, and appeared to lead to something which 

 may have affected the results considerably, all that is necessary to know now being 

 that the comparison of temperatures given by this circuit is the same as that by tin- 

 previous one. 



To effect this comparison an experiment was made with saturated steam under 

 known pressures flowing through the apparatus, the orifice plate having been removed 

 for this purpose in precisely the same way as described in correcting the thermo- 

 meters. When the conditions were steady, each of the two circuits from the 

 junctions in the steam were brought in turn into circuit with the galvanometer. The 

 mean of the actual readings on the thermometer in the oil bath given by the 

 two circuits were 248 '05 by the new circuit and 246 '95 by the old one, and again 

 248'0 by the new circuit, the difference being about 1'1 F. 



Thus it appears that when the results given by the new circuit are corrected in tlic 

 same manner as were those with the previous circuit, the corrected results do not 

 differ by an amount so great as the natural error of experiment, and above all it is 

 shown that experiments made with the junction immersed in the steam in the 

 horizontal portion of the channel, and therefore not in a position directly opposite the 

 orifice, as is the case with the old junction, would give results which within the limits 

 of experimental accuracy do not differ from those already obtained with the old 

 circuit. 



In the last experiment made (No. 28), since the temperatures were here very high, 

 and the flow of steam very great, after the completion of the experiment with the aid 

 of the old circuit, the new circuit was brought into play to see if this same difference 

 existed at higher temperatures. The actual readings taken with the old circuit at the 

 pressure 1575 Ibs. being 310'25, and with the new circuit 311'3, the difference 

 l>eing r05, which is practically the same as that found to exist between the observa- 

 tions given by the two junctions at lower temperatures. Hence, as has been shown, 

 the same results would have been obtained had either of the two positions in the 

 steam channel been initially chosen to plan the thermo-j unction. 



SECTION XIV. On the Transference of Heat across the Orifice Plate. 



In the experiments made with high initial temperatures the difference of tempera- 

 ture on the two sides of the orifice plate sometimes amounted to over 50, and if only 

 a small quantity of steam is flowing through the channel the heat transferred from 

 one side to the other of the orifice may become relatively very important. 



In order to calculate the maximum difference of temperature caused by this 

 transference, the author took an hypothetical case in which the difference of tempera- 

 ture on the two sides of the orifice plate was 50 F., and the quantity of steam 1 Ib. 

 in 12 minutes, the effect of this combination of circumstances being greater than any 

 actually experienced in the experiments. Thus the quantity of heat conducted 



