Mr. Combe observes, in his introduction to his notes on the United States of North America, p. xi.--'I was told that a _certain person_ boasts of having given Miss Martineau erroneous information for the purpose of leading her into mistakes; and another in Philadelphia assures his friends that he "crammed" Capt. Marryatt with old "Joe Millers," which the Capt. embodied into his books as facts illustrative of American manners. This seems to be a case in which some uncertainty must ever exist as to the value of the facts collected by travellers. They cannot observe all, or test half that they do observe. They must rely on testimony. But they might do this--They might tell us precisely the kind of authority they followed, and then the reader could form some opinion of the value of what was communicated. Had Miss Martineau and Captain Marryatt given the name and addresses of their informants, the latter would now be punished by being infamously known throughout Europe; and all future travellers warned from them--and all future informants warned by their example. Where informants cannot be mentioned by name and address, the chances are, they cannot be trusted. When first connected with public proceedings, I found myself made the depository of innumerable bits of scandal, and ominous reports of public characters. To all who told me anything, if I attached importance to it, I made it a rule to ask--'May I mention it to the party with your name?' 'O, no, I would rather not,' was the common reply. To all written communications answer--'Please add your name and address--and may I publish them if occasion requires?' 'O, no, don't,' would be the general injunction. Thus I found that huge reports, inflated as balloons, shrunk like them when pricked by the pin of a question--'Will you answer for it?' Thus I saved myself from being imposed upon by, or being the retailer of, reports for which the originator or relator would not or could not vouch.