Let us remark, firstly, that the space of time in question is much more limited still. M. Loiseleur only made use of documents already published. But, November 18, 1681, Louvois, in a despatch as yet unprinted, says to Saint-Mars, with reference to his prisoners: “The King approves of your choosing a doctor to visit your prisoners, and of your employing the Sieur Vignon to confess them once a year.” From this it would appear that a new prisoner was confided to Saint-Mars, between September 20 and November 18, 1681; but, as we have already said in the last chapter, there is nowhere any trace of this prisoner, this so-called spy. On the other hand, for the despatch of September 20, 1681, to have the meaning which M. Loiseleur attributes to it, one of the two prisoners of the lower Tower must have died some days previous to September 20, since at this date only one prisoner is spoken of. Of this death or disappearance we have no proof or even trace. Thus the whole argument rests upon this single despatch, of which M. Loiseleur not only makes use in order to prove that a new prisoner had been confided to Saint-Mars, but from which he also deduces that one of the prisoners previously confined had disappeared. This single despatch thus standing alone, and completely unsupported, would be far from being sufficient to establish this theory. Nevertheless, it is essential to discover its true meaning, so as to leave no doubt in the reader’s mind, and to make every part of our demonstration clear and plain. I acknowledge having spent a considerable time in thinking over this despatch, which was contradicted by all the others, which suited no theory, and which was nevertheless authentic and very exactly reproduced, since I went several times to read the draft of it, at the Archives of the Ministry of War. Even if it had possessed the meaning that M. Loiseleur attributed to it, it would not have destroyed my conclusions at all, since the proofs which I had furnished of the obscurity of the Exiles prisoners were also applicable to this new prisoner brought between November and September, 1681, and because the superior importance of the prisoners afterwards taken from Pignerol to the Isles Sainte-Marguerite, would not have been demonstrated any the less clearly by the despatches which I am about to quote. But it was repugnant to me to leave a single point obscure; and after much reflection, and after having been for a long while of M. Loiseleur’s opinion, although nothing outside of this despatch justifies his interpretation, I believe that I have discovered its true meaning.