Page:Proceedings of the Royal Society of London Vol 69.djvu/66

58 obviously wrong. Otherwise the actual means were used so that the error might be the more readily estimated. Had I been able to obtain printed uniform patches I have no doubt the results would have been better. But even these can leave no doubt of the approximate truth of the law.

Curve No. 20. I. A very faint rubbing.

Scale.

Aperture.

Average.

Corrected from curve.

White.

Per cent.

Log of ratio.

10

10 -0 10 -5

10-2

81-2

1 12-4 ! -0-906

15

16-0 - 14-5

15-2

85-4

17'7

-0 -752

20

27-0 - - 27-0

27 -0 87 -0

31-0

-0-508

25

35-5 - 35-0

35 -2 86 -6

40-6

-0-392

30

43 -0 43 -0

43-0

83-5

51-6

-0-288

35 43-0 46-0 45 41 -6

77-2

57-8

-0-238

40 45-0 - - 46-0 i 45-5

69-3

65-7

-0-182

45

50-0 53-0 48-50

50 -0 64 -4

70-0

-0-155

50

38 -0 42 -0 39 '0 39 '6 | 57 '3

69-1

-0 -160

55

38-5 38-0 38-2

54-7

69-7

-0-157

Curve No. 21. II. A little darker than I.

Scale.

Aperture.

Average.

Corrected from curve.

White.

Per cent.

Log of ratio.

10

4 -5 _ 4-0

4-2

81-2

5-2

-1-384

15

6 -0 6-5

6-7

.

85-4

7-8

-1-107

20

10 -5 9 -5

10-0

87-0

11-5

-0-939

25

18-0 - 16-5

17-2

86-6

20-0

-0-699

30

23-0 - - 24-0

23-5

83-5

28-1

-0-551

35

27 5 26-5 28-0

27-3

77-2

35-5

-0-450

40

32-0 - - 32-0

32-0

69-3

46-0

-0-337

45

33 -0 35 -0 33 '0

33-6

64-4

52-2

-0-282

50

27 -0 28 -0 32 -0

29-0

32

57-3

55-8

-0-254

55

30 -0 29 -5

29-8

54-7

53-4

-0-272