CONTINUOUS ELECTRIC CALORIMETRY. 
75 
standards are given in Table VII. (Barnes). The mean of the results may be stated 
as follows :•—- 
No. 4086 at 21°"74 C. = Mean of 10 Pt-Ag ohms - -00054 at 21°"59 C. . (3). 
The coils were also compared about the same time by Mr. Fraser at a lower 
temperature. The value of No. 4086 in terms of the mean deduced from this second 
comparison is :—- 
No. 4086 at 14 c, 8 C. = Mean of 10 Pt-Ag ohms -f ’00153 at 13°*6 C. . (4). 
The details of this comparison are given in Table VIII. (Barnes). 
If the mean values of the 10 Pt-Ag ohms at 21 0, 6 and 13°"6 C., respectively, are 
deduced from the certificated value of the mean at 16°'5 Cl, adopting '000254, the 
value taken by Barnes as the temperature coefficient, we obtain the following values 
for No. 4086 :— 
No. 4086 at 21 0, 74 C. = 1 '00037 ohms (from standards at 21°'6 C.) . (5). 
„ „ „ 14°'8C. =1'00041 „ „ „ 13°'6 C.) . (6). 
These results are evidently inconsistent with the value + "000018 for the tempera¬ 
ture coefficient of No. 4086. The individual observations, however, are too consistent 
to admit of the supposition of an error of the order of two parts in 10,000 in the 
mean of the comparisons. It seems more likely that the value "000254 assumed for 
the temperature coefficient of the Pt-Ag standards is too small. In any case, if we 
take the mean of the two comparisons, we shall obtain a result which is nearly 
independent of the value assumed for the temperature coefficient. We thus 
obtain :— 
No. 4086 at 18°"4 C. = 1 "00039 ohms (Pt-Ag standards at 17°"6 C.) . (7). 
It happens that the temperature 18°"4 C. is precisely that at which the value of 
No. 4086 is required for the comparisons with the mica current-standards. 
(8.) Comparisons at the National Physical Laboratory. 
It appeared practically certain from these comparisons that the value of No. 4086 
had increased since the date of the original certificate by nearly "00060 ohm. For 
further verification No. 4086 was sent over in a box from Montreal to Cambridge, 
and thence to Kew, to be compared with the original standards. Mr. Glazebrook 
very kindly undertook the comparison himself, with the following results :— 
