based upon the accusations of "good witches," upon the statements of the dying, or upon the charges of those who had suffered ill after threats, he thought ought to be used with great caution. It is evident that Perkins--though he doubtless would not have admitted it himself--was affected by the reading of Scot. Yet it is disappointing to find him condoning the use of torture[8] in extreme instances.[9] A Cambridge man who wrote about a score of years after Perkins put forth opinions a good deal farther advanced. John Cotta was a "Doctor in Physicke" at Northampton who had taken his B. A. at Cambridge in 1595, his M. A. the following year, and his M. D. in 1603. Nine years after leaving Cambridge he had published _A Short Discoverie of the Unobserved Dangers_, in which he had devoted a very thoughtful chapter to the relation between witchcraft and sickness. In 1616 he elaborated his notions in _The Triall of Witchcraft_,[10] published at London. Like Perkins he disapproved of the trial by water.[11] He discredited, too, the evidence of marks, but believed in contracts with the Devil, and cited as illustrious instances the cases of Merlin and "that infamous woman," Joan of Arc.[12] But his point of view was of course mainly that of a medical man. A large number of accusations of witchcraft were due to the want of medical examination. Many so-called possessions could be perfectly diagnosed by a physician. He referred to a case where the supposed witches had been executed and their victim had nevertheless fallen ill again.[13] Probably this was the case of Mistress Belcher, on whose account two women had been hanged at Northampton.[14] Yet Cotta believed that there were real witches and arraigned Scot for failing to distinguish the impostors from the true.[15] It was indeed, he admitted, very hard to discover, except by confession; and even confession, as he had pointed out in his first work, might be a "meane, poore and uncertain proofe," because of the Devil's power to induce false confession.[16] Here the theologian--it was hard for a seventeenth-century writer not to be a theologian--was cropping out. But the scientific spirit came to the front again when he made the point that imagination was too apt to color observations made upon bewitched and witch.[17] The suggestion that coincidence explained many of the alleged fulfillments of witch predictions[18] was equally in advance of his times.