Occasionally we receive complaints because a paper we have published contains material which is not new and does not make reference to the earlier work. This is indeed unfortunate, since a physicist's principal reward for his achievements is recognition by his colleagues. To see one's own work overlooked and forgotten is a painful experience but it happens often, especially when the subject is theoretical speculations. In order not to hurt anyone else's feelings, I shall cite as an illustration a case of my own experience. In the first paper I ever published I tried to establish empirically that the doublets in atomic spectra obeyed the same relativistic formula that was used by Sommerfeld for x-ray doublets. I did not consider this a mere gamble, but believed that I had made a true contribution. My teacher, Ehrenfest, must have thought differently; but in order not to discourage a young beginner, he made me write a short note which appeared in Die Naturwissenschaften of 9 December 1921. A long paper was published in the then already obscure "Archives Néerlandaises des Sciences Exactes et Naturelles," printed in Holland in the French language as if to make sure that almost no one would read it. Ehrenfest also invited me to attend henceforth his famous Wednesday evening colloquia, and to present my paper there. In the fall of 1924 Dirk Coster reported in the colloquium on articles by Landé and by Millikan and Bowen. I meekly remarked that I had talked about the very same ideas almost three years earlier. I did not appreciate that in the meantime a considerable amount of new spectroscopic data had been obtained, especially by Millikan, which fully supported the relativistic formula, whereas my guess had been based on totally insufficient evidence. In subsequent papers Landé was very kind and mentioned that "indications" for the validity of a relativistic doublet formula were contained in my early note. Millikan was not so generous. Late in 1925 Ehrenfest wrote him about the introduction of electron spin, which finally had furnished a sound basis for the doublet formula, and noted how nice it was that one of the proponents of the spin idea had also been the first to propose that formula. Ehrenfest told me that Millikan had reacted rather unfriendly to this remark, but refused to show me his letter. Thus, it seems to me that in spite of the great changes in physics the attitudes and sensitivity of physicists have not changed in all those years and probably never will. In a book review in this month's Bulletin of the Atomic Scientists, Hedwig Born considers publishing of half-baked ideas unethical, since the author gambles without risking a loss. If his idea is later proven to be correct he claims priority and enhances his reputation; if it is proven false he "offered it only as a hypothesis" and suffers no discredit. Even the best referee cannot always remember every paper in his own specialty. A new discovery or idea will be remembered without difficulty if it is accepted at the time it is announced. As long as it is still merely a suggestion or a speculation, it seldom leaves a lasting impression. In such cases the same "new" idea can be rediscovered several times. The questions arise whether the earlier proponents were far-sighted geniuses or merely lucky guessers and also whether their speculations influenced later work. It is obvious that these questions can rarely be settled objectively. Therefore, in most cases, credit should be granted only when the discoverer has waited until other developments have furnished arguments which make the idea finally acceptable.