"Where doubt arises," says Starkie (_Ev._ i. 586), "from circumstances of an apparently opposite and conflicting tendency, the first step in the natural order of inquiry is to ascertain whether they be not in reality reconcilable, especially when circumstances cannot be rejected without imputing perjury to a witness; for perjury is not to be presumed, and in the absence of all suspicion that hypothesis is to be adopted which consists with and reconciles all the circumstances which the case supplies." (See also _Starkie_, i. 578, 582.) Take the familiar case of the taxing when Cyrenius was governor of Syria. Luke ii. 2. Everybody knows how confidently it was asserted that Luke was in error because Cyrenius' government of Syria was several years later than Luke makes it; equally, every one knows how that difficulty was met by the supposition, made almost a certainty, that Cyrenius was twice governor of Syria--once at the time in question, and once later. Even if the supposition were not as probable as it is, if there were no other way of solving the difficulty, we should be justified by the principle of law in assuming it rather than to assume that a witness as intelligent as Luke, and with his opportunities of knowledge and with no motive for misstatement, should either wilfully or carelessly have made so gross an error. Here the rule fits perfectly: "In the absence of all suspicion, _that hypothesis is to be adopted which consists with and reconciles all the circumstances which the case supplies_." In regard to certain objections to the Mosaic record--for example, the improbability of the desert sustaining the host of the Israelites: we select this as an example of a mass of like objections--Dean Stanley, while holding in general to the historic fact, says the recorded miracles do not meet the difficulty and we have no right to add to them; for "if we have no warrant to take away, we have no warrant to add." If by this he meant we have no right to add to the inspired word _as a part of it_ what is not in it, he is quite correct; but if he meant, as he evidently did, that we have no right to make a reasonable supposition to explain an apparent difficulty of the word, no utterance can be more groundless. He might as well object that Moses could not possibly have led the Israelites through the desert forty years because no man could do that without sleeping, and the record does not say that Moses slept during all that time, and "we have no warrant to add" to the record.