Few cases will occur in which it is not possible to procure evidence of the kind, when diligently sought for. It is often too very decisive in its operation on judicial proceedings. In the case of Margaret Wishart tried at the Perth Spring Circuit in 1827 for poisoning her blind sister, a man who lodged with the prisoner and cohabited both with her and with the deceased, appeared at first from general circumstances to be implicated in the crime. He had left the house, however, on the morning of the day before that on the evening of which the deceased took ill; and he did not return till after her death. Now her illness commenced suddenly and violently; and arsenic was the poison which caused it.[125] It was quite clear, therefore, that the poison could not have been administered, at least in a dangerous dose, so early as the day before she was taken ill; and such I stated to be my opinion, on a reference from the Lord Advocate. The evidence being also otherwise insufficient, the man was set at liberty. In the case of Mrs. Smith tried here in February of the same year, this branch of the evidence was made the subject of question under more doubtful circumstances. The deceased certainly died of poisoning with arsenic, and the prisoner was strongly suspected of being the poisoner for many reasons, and among others because, on the evening before the morning on which the deceased took ill, the prisoner gave her in a suspicious manner a white-coloured draught. Here the possibility of the draught having been the cause of the symptoms must be admitted. But as they did not appear for eight hours after the draught was taken, I stated in my evidence that it was improbable the dose, if it contained arsenic at all, contained a quantity sufficient to cause the violent symptoms and death which followed.[126]