NOTES ON PLATE EXPOSURES. 189 
Other plates. The use of the screen in conjunction with 
panchromatic plates would appear to give interesting 
results—one is sorry that an exposure was not made on the 
16 foot coronagraph by this method, for the inner corona. 
The one and three seconds were definitely underexposed, 
except for inner corona—probably a 20 seconds exposure 
f 6°3 would have been best, although there was very little 
difference between the 10 and 30 seconds. The images 
were quite sharp, showing no sign of movement even in 
the 60 seconds exposure, and permitting enlargement up 
to ten diameters, beyond which the grain of the plate spoils 
all detail. 
The box camera plates were not so satisfactory—the 
images were not so sharp and streamer definition not so 
good, though again extending out to three solar diameters. 
Considering the sharpness of the lunar image in trial plates, 
it is considered that possibly some slight movement of the 
slide holder with respect to the focal plane took place after 
rehearsal, and prior to the eclipse. The 30 seconds exposure 
f 13 was quite satisfactory and would, with the aeroplane 
camera results, seem to point to an exposure of the order 
20 seconds f 8 as being the best for outer corona work. 
Having now a good collection of negatives, it was required 
to obtain prints showing everything that was included in 
the plates. It is well known that the best detail is seen 
in the original negative itself—it is not possible to produce 
an uncontrolled print in which so much can be seen. For 
the examination of the plate, therefore, or for exhibition, 
it is best to print off transparencies, on the slowest and 
heaviest emulsion obtainable. 
The transparencies consisting of silver deposits in gela- 
tine, the thickness of the silver layer being a function of 
the incident light value, and being viewed by transmitted 
