As early as 1819 Prof. Benjamin Silliman declined to publish contributions from him in the "American Journal of Science,"[255] and returned him his MSS. Dr. Gray strongly intimates that Rafinesque's assertions on scientific matters were at times intentionally false, as when he said that he had seen Robin's collection of Louisiana plants in France, whereas that botanist never prepared dried specimens; and the like. I felt early in this investigation that Rafinesque's assertions were, therefore, an insufficient warranty for the authenticity of this document. As I failed in my efforts to substantiate them by local researches in Kentucky and Indiana, I saw that the evidence must come from the text itself. Nor would it be sufficient to prove that the words of the text were in the Lenape dialect. With Zeisberger and Heckewelder at hand, both of whose works had been years in print, it were easy to string together Lenape words. But what Rafinesque certainly had not the ability to do, was to write a sentence in Lenape, to compose lines which an educated native would recognize as in the syntax of his own speech, though perhaps dialectically different. This was the test that I determined to apply. I therefore communicated my doubts to my friend, the distinguished linguist, Mr. Horatio Hale, and asked him to state them to the Rev. Albert Anthony, a well educated native Delaware, equally conversant with his own tongue and with English. Mr. Anthony considered the subject fully, and concluded by expressing the positive opinion that the text as given was a genuine _oral_ composition of a Delaware Indian. In many lines the etymology and syntax are correct; in others there are grammatical defects, which consist chiefly in the omission of terminal inflections. The suggestion he offered to explain these defects is extremely natural. The person who wrote down this oral explanation of the signs, or, to speak more accurately, these chants which the signs were intended to keep in memory, was imperfectly acquainted with the native tongue, and did not always catch terminal sounds. The speaker also may have used here and there parts of that clipped language, or "white man's Indian," which I have before referred to as serving for the trading tongue between the two races. This was also the opinion of the Moravian natives who examined the text. They all agreed that it impressed them as being of aboriginal origin, though the difference of the forms of words left them often in the dark as to the meaning. This very obscurity is in fact a proof that Rafinesque did not manufacture it. Had he done so, he would have used the "Mission Delaware" words which he found in Zeisberger. But the text has quite a number not in that dialect, nor in any of the mission dictionaries.