What was the exact value set upon evidence procured by torture it would be difficult at this day to determine. We have seen above that Augustus pronounced it the best form of proof, but other legislators and jurists thought differently. Modestinus affirms that it is only to be believed when there is no other mode of ascertaining the truth.[1445] Adrian cautions his judges not to trust to the torture of a single slave, but to examine all cases by the light of reason and argument.[1446] According to Ulpian, the imperial constitutions provided that it was not always to be received nor always rejected; in his own opinion it was unsafe, dangerous, and deceptive, for some men were so resolute that they would bear the extremity of torment without yielding, while others were so timid that through fear they would at once inculpate the innocent.[1447] From the manner in which Cicero alternately praises and discredits it, we can safely assume that lawyers were in the habit of treating it, not on any general principle, but according as it might affect their client in any particular case; and Quintilian remarks that it was frequently objected to on the ground that under it one man’s constancy makes falsehood easy to him, while another’s weakness renders falsehood necessary.[1448] That these views were shared by the public would appear from the often quoted maxim of Publius Syrus—“Etiam innocentes cogit mentiri dolor”—and from Valerius Maximus, who devotes his chapter _De Quæstionibus_ to three cases in which it was erroneously either trusted or distrusted. A slave of M. Agrius was accused of the murder of Alexander, a slave of C. Fannius. Agrius tortured him, and, on his confessing the crime, handed him over to Fannius, who put him to death. Shortly afterwards, the missing slave returned home. This same Alexander was made of sterner stuff, for when he was subsequently suspected of being privy to the murder of C. Flavius, a Roman knight, he was tortured six times and persistently denied his guilt, though he subsequently confessed it and was duly crucified.[1449] A somewhat similar case gave Apollonius of Tyana an opportunity of displaying his supernatural power. Meeting in Alexandria twelve convicts on their way to execution as robbers, he pronounced one of them to be innocent, and asked the executioners to reserve him to the last, and, moreover, delayed them by his conversation. After eight had been beheaded, a messenger came in hot haste to announce that Phanion, the one selected by Apollonius, was innocent, though he had accused himself to avoid the torture.[1450] A curious instance, moreover, of the little real weight attached to such evidence is furnished by the case of Fulvius Flaccus, in which the whole question turned upon the evidence of his slave Philip. This man was actually tortured eight times, and refused through it all to criminate his master, who was nevertheless condemned.[1451] The same conclusion is to be drawn from the story told by St.