I considered the matter for some time before replying. “Not with any confidence,” I answered at length, “unless it was, as people have suspected, to take her own life.”

“Is there no other reason that someone might go to a quiet, secluded spot, but to commit suicide?” queried my friend in a sceptical tone.

“To enjoy the peace of the countryside,” I suggested, without any great conviction. “To reflect upon one’s life, perhaps.”

“Perhaps. But there are other, more commonplace reasons, which you seem determined to overlook. Do you recall the lunchtime of last Friday?”

“Certainly,” I replied in surprise.

“You paid a visit then to the Criterion Bar. Why was that?”

“I had arranged to meet someone there.”

“Precisely! Is it not at least possible, Watson, that Sarah Dickens went to the Willow Pool on the day of her death because she, too, had arranged to meet someone there?”

“Yes, I suppose it is.”

“Thank you. Turning now from Sarah Dickens’s purpose in going to Jenkin’s Clump that day to the situation of her lifeless body when discovered, can you recall where it is said her body was found?”

“In the water, near the bank, on the far side from the footpath; that is, the northern bank, where the brambles grow thickly.”

“Very well. From which part of the bank, then, do you think she entered the water?”

“The north side, surely, close to where her body was found.”

“Why do you think she had gone round to that side of the pool?”

I hesitated. “It was thought that she had gone round to that side in order to pick blackberries,” I replied after a moment, “but that does not now appear so likely.”

“Quite so, especially when you consider that she had left her basket behind, on the south side of the pool. And nor had she returned to the north side to retrieve something she had dropped when picking blackberries earlier, for she had picked no blackberries earlier – the basket was empty. Nor can it be suggested with any plausibility that she took herself round to the north side of the water in order to stand and read her note there: it is a difficult, prickly spot, and there is barely space between the brambles and the water for anyone to stand. No one would choose to go there except to pick blackberries, and that, as we have seen, Sarah Dickens was not doing.”

“Perhaps,” I suggested, “her reason for going to the Willow Pool was to retrieve the note, which she realized she had dropped there on a previous visit. She may have seen it in the brambles from across the pool, left her basket by the footpath on the south side of the pool, and made her way round to the north side to try to reach the note.”

“Ingenious, Watson!” cried Holmes with a chuckle. “That is, indeed, on the face of it, a possibility. Let us continue the hypothesis a little longer, then. What do you suggest happened next?”

“The most likely sequence of events is this,” I began when I had considered the matter for a moment: “that she was stretching to try to reach the note, when she lost her balance, and perhaps her footing, too, and fell into the water. There she hit her head upon a submerged stone, as was suggested at the original inquest, lost consciousness and drowned.”

“Capital!” cried my friend, clapping his hands together.

“You think there is some truth in that suggestion, then?” I asked, pleased that my hypothesis appeared to meet with his approval.

“No. None whatever.”

“What!”

“I simply meant that you have presented your hypothesis cogently. It illustrates very clearly how convincing a hypothesis can be, even though quite fallacious, when it is derived from only a selection of the evidence, rather than from all of it.”

“Pray, let us hear your own view, then,” I retorted – irritated, I admit, by his tone of superiority.

“First of all,” he replied, “on your hypothesis – and that which is implicit in the verdict of the original inquest – the girl would not have drowned had she not lost consciousness, and she would not have lost consciousness had she not struck her head upon a stone.”

“That is so,” I concurred. “Indeed, that seems an obvious inference.”

“Well,” said he, “it might well have been an obvious inference had there been any stones in the water. But there are none there, Watson, as I ascertained this afternoon. I am, I take it, the first person who has troubled himself to verify the assumption.”

“Perhaps, then, she struck her head on the bottom of the pool,” I suggested, “or on the bank as she slipped into the water.”

“The bottom of the pool is soft and sandy, the bank grassy. Neither would have produced the bruise described by the local medical officer at the inquest.”

“What then?”

“Well, it is your theory, old fellow. How do you account for the girl’s drowning, if she did not knock herself unconscious?”

“We know from Mr Yarrow,” I replied after a moment, “that her hair was tangled in the brambles that drooped into the water. Perhaps, then, after she fell in she became hopelessly entangled, with her head under the water and drowned in that way.”

“The water is no more than three feet deep in that part of the pool,” returned Holmes. “She could easily have stood up, out of the water.”

“But if her hair was entangled?”

“Brambles are undoubtedly tough and troublesome,” said Holmes, “but not so tough that a normal adult cannot overcome then. Besides, if the girl had struggled to free herself from prickly brambles, her hands would have been covered with scratches, and her scalp and face, too; yet the medical officer’s report, read out at the inquest, specifically stated that there were no marks upon the body whatever, other than the bruise on the side of the head.”

“Very well,” said I. “I concede that you have disproved the view that Sarah Dickens’s death was the result of mischance. The verdict of the inquest was wrong. But what of the possibility that she took her own life?”

“How might she have done that, do you suppose?” asked Holmes.

“I cannot pretend to be an expert in such matters, but I suppose she would have walked into the water until she was out of her depth, swallowed water and thus drowned.”

“And then?”

“I am not sure what you mean. Presumably her body would have remained floating in the water until it was discovered.”

“If we accept this hypothesis for a moment,” said Holmes, “where, then, must the girl have entered the water?”

“There are only two places, realistically speaking,” I returned. “Most of the circumference of the pool is thickly overgrown with brambles, briars, nettles and so on, and can thus be discounted. The two possible places are the flat, open area on the south bank, where we sat on the fallen tree, and the narrow strip by the brambles on the north bank, which we have spoken of already. Of the two, the former is by far the more likely. If the girl did not go round to the north bank in order to pick blackberries, or to retrieve that piece of paper, then there was no reason for her to go round there at all. It would be much easier, as well as more direct, for her to enter the water from the south bank.”

“Very good. And do you recall the position of the body when found, according to the testimony of Mr Yarrow?”

“He said that it was a couple of feet from the bank on the north side, a little higher up the pool than the spot where one could pick blackberries.”

“Did you observe the current in the pool?”

“I did. I observed your experiments with the sticks this morning. It runs, of course, from west to east, the direction of the stream that feeds and drains the pool.”

“How do you suggest, then, that the body of Sarah Dickens was found on the other side of the pool, and upstream of the spot from where she must have entered the water?”

I considered the matter for a moment. The point had not, I confess, occurred to me before.