And let us suppose that when the Dutch stem of Schermerhorn and the French stem of de Lancey come into view in our family tree, we find Joneses again, and--yes, a little research proves that these Joneses also descended from the emigrant, Stephen Jones, the ancestor of our maternal grandfather, Colonel William Jones. The Jones stock is a fine brand, and three strains are none too many, but their appearance subtracts two more surnames from the theoretical number. Furthermore, while we may be able to find our way back from generation to generation with almost ridiculous ease in some cases, such luck is usually too good to last. It is a rare vein which yields family connections at every stroke of the genealogical spade, and one such line may have to console us for a number which we mine slowly and painfully, and for some others which yield no results whatever beyond a certain point. In truth, most old American families pan out fairly well, with here and there a golden nugget of peculiar lustre, or a diamond of the first water; but we are seldom troubled by finding more of this wealth than we are able to handle. In making the investigation, we should aim to collect data for a very full account of each ancestor, with a portrait, autograph, the history of his possessions, photograph of the homestead, his old letters, his Bible and will--in fact, any and all materials which picture clearly his character and affairs. When we have finished collecting, our accumulations are worked up into monographs on each one of the lines traced, each monograph enriched by illustrations and accompanied by an appendix in which we exhibit in full the documents and extracts constituting the proofs of the descent. We recommend that each monograph be introduced by a chart, exhibiting the pedigree from the earliest known progenitor down to the person whose ancestry is the subject of the book. This adds a valuable feature, and makes the whole line clear at a glance. After all the monographs are completed, they should be arranged together for publication in one volume. If expense is not much of an object, it is especially interesting to prepare for one's own library one copy of the edition printed, sumptuously bound and enriched with original documents, or certified copies of them,--old prints, silhouette portraits and other illustrations gathered solely for that copy. In fact, some people may prefer to limit the edition to this one copy. These ideas may be followed in the Grafton plan of genealogy with brilliant results. A proper method of research, with the necessary means at its disposal, should result in the accumulation of an abundance of interesting illustrative matter for such a book.