134 
MR. W. GEOFFREY DUFFIELD ON THE 
points, as these indicated their most intense portions and enabled a more accurate 
setting of the cross-wires to be made. Two parallel threads in the'eye-piece gave 
more consistent results than cross-wires. 
As has already been stated, two sets of photographs, A and B, have been taken at 
pressures varying from 1 to 101 atmospheres (absolute). 
Set A was measured first by the writer, and then by two assistants. Each observer 
obtained six readings of the displacement, making the total number of readings for 
each line 18. This number was frequently exceeded. For a second plate taken at 
the same pressure, 12 readings of each line were usually considered sufficient. The 
mean value of the shift of each line at each pressure was therefore obtained from at 
least 30 readings of the displacement. For Set A, 20,000 determinations of the 
displacements have been made. 
Set B was measured by the writer alone, except in a few cases when separate 
observation seemed desirable, and the figures given in the table are the means of 
ten measurements of each displacement. The plates used for this set were Imperial 
Flashlight, and the Developer the Imperial Pyro-Metol Standard—a combination 
which was more successful than that previously used for Set A ; the photographs 
were free from developer-fog, and the lines consequently measurable with much 
greater ease. The total number of readings for the second set is about one-third of 
the number made for the first set, but for the reasons given above, and also because 
experience is essential for the accurate determination of the positions of the most 
intense portions of the lines under pressure, the results given for the second set are 
considered more satisfactory. 
The following photographs constitute the two sets :— 
