As a thief in the night

by R. Austin Freeman

“But complete as the proof was, I wished, if possible, to make it yet more complete: to associate Barbara still more definitely with the crime. In the case of Monkhouse, it was clear that the poisoning always occurred when she was absent from home. But this was not so clear in the case of Stella. Your diary showed that Stella’s relapses coincided pretty regularly with Barbara’s absences; but it was not certain (though obviously probable) that the relapses coincided with the periods of poisoning. If it could be proved that they did coincide, that proof would furnish corroboration of the greatest possible weight. It would show that the two cases were parallel in all respects.

“But could it be proved? If the tress of Stella’s hair had been at my disposal, I had no doubt that I could have decided the question. But the tress was yours, and it had to be preserved. Whatever was to be done must be done without destroying or injuring the hair, and I set myself the task of finding some practicable method. Eventually, I decided, without much hope of success, to try the X-rays. As arsenic is a fairly dense metal and the quantity of it in the deposits quite considerable, it seemed to me possible that it might increase the density of the hairs at those points sufficiently to affect the X-ray shadow. At any rate, I decided to give the method a trial.

“Accordingly, Polton and I set to work at it. First, in order to get the densest shadow possible, we made the tress up into a close cylinder, carefully arranging it so that all the cut ends were in exactly the same plane. Then we made a number of graduated exposures on ‘process’ plates, developing and intensifying with the object of getting the greatest possible degree of contrast. The result was unexpectedly successful. In the best negative, the shape of the tress was faintly visible and was soon to be crossed by a number of perfectly distinct pale bands. Those bands were the shadows of the deposits of arsenic. There could be no doubt on the subject. For, apart from the fact that there was nothing else that they could be, their appearance agreed exactly with what one would have expected. Each band presented a sharp, distinct edge towards the tips of the hairs and faded away imperceptibly towards the roots. The sharp edge corresponded to the sudden appearance of arsenic in the blood when the poisoning began. The gradual fading away corresponded to the period of elimination when the poisoning had ceased and the quantity of arsenic in the blood was becoming less and less from day to day.

“Now, since hair grows at a known, uniform rate, it was possible to convert the distances between these arsenical bands into periods of time; not with perfect exactness, because the rate of growth varies slightly in different persons, but with sufficient exactness for our present purpose. As soon as I looked at those bands, I saw that they told the whole story. But let us follow the method of proof.

“Assuming the rate of growth to be one fiftieth of an inch in twenty-four hours—which was probably correct for a person of Stella’s age—I measured off on the photograph seven inches and a quarter from the cut ends as representing the last year of her life. Of course, I did not know how close to the head the hair had been cut, but, judging by the bands, I assumed that it had been cut quite close to the skin—within a quarter of an inch.”

“I happen to know that you were quite right,” said I, “but I can’t imagine how you arrived at your conclusion.”

“It was quite a simple inference,” he replied, “as you will see, presently. But to return to the photograph. Of the measured space of seven inches and a quarter I took a tracing on sheet celluloid, marking the sharp edges of the bands, the points at which the fading began and the points at which the band ceased to be visible. This tracing I transferred to paper ruled in tenths of an inch—a tenth of an inch representing five days—and I joined the points where the fading began and ended by a sloping line. I now had a diagram, or chart, which showed, with something approaching to accuracy, the duration of each administration of arsenic and the time which elapsed between the successive poisonings. This is the chart. The sloping lines show the fading of the bands.”

He handed me a paper which he had just taken from a drawer and I looked at it curiously but with no great interest. As I returned it after a brief inspection I remarked:

“It is quite clear and intelligible, but I don’t quite see why you took the considerable trouble of making it. Does it show anything that could not be stated in a few words?”

“Not by itself,” he replied. “But you remember that I mentioned having made two other charts, one showing the fluctuations in Stella’s illness and the other showing Barbara’s absences from home during the same period. Here are those other two charts; and now, if you put the three together, your eye can take in at a glance a fact of fundamental importance; which is that the relapses, the absences and the poisonings all coincided in time. The periodicity is strikingly irregular; but it is identical in all three charts. I made these to hand to the jury, and I think they would have been quite convincing, since any juryman could check them by the dates given in evidence, and by inspection of the radiograph of the hair.”