Mr. Sinclair remarks that “when in these drawing tests there has been
anything [that is, in his drawings] indicating fire or smoke she has
‘got’ it, with only one or two failures out of more than a dozen cases.”
This would mean a much larger ratio of success for the drawings so
characterized than that for the total number of drawings. Mr. Sinclair
accounts for this by the fact that his wife, owing to terrifying
incidents in her childhood, is exceedingly sensitive to the thought of
fire and given to taking unusual precautions. Readers will probably
agree that this is a plausible and sensible theory. I propose to
tabulate _all_ such tests, including the original drawings significant
of light.