II. THE TRADITION OF ANCIENT 

 AND MEDIAEVAL SCIENCE 



When men of science become interested in the history of science, 

 their interest is generally focussed upon the immediate past, or what we 

 might call "modern" science — however this may be defined. They may 

 choose to begin it with the western reinvention of typography ( c. 1450 ) , 

 or with Copernicus or Ves alius (1543), or with Kepler (1609-19) and 

 Galileo (1632-38), or with Newton (1687), or with Volta (1800), or 

 with the introduction of astrophysics, or radioactivity, or later still. 

 Each of these limits can be justified, and one is as good as another. Al- 

 most every man of science, whether he be historically minded or not, is 

 obliged to do a certain amount of retrospection, because his own investi- 

 gations bring him face to face with the work of some predecessor, or 

 because of academic conventions. The historical difficulties of such 

 superficial retrospect are not great, the sources are easily obtainable, the 

 chronological basis is relatively easy to establish. The fundamental 

 questions "When did that happen? where?" are easy to answer. The 

 questions "why?" and "how?" are more difficult of course, yet they are 

 still comparatively easy for late periods. Men of science whose retro- 

 spective insight does not go much deeper than the last century have few 

 chronological troubles to speak of ^ and no idea of the vicissitudes of 

 tradition. Consider Oersted's famous paper of 1820 which is the foun- 

 dation of electromagnetism; originally written in Latin, it was promptly 

 translated into French, Italian, German, English, and Danish, and within 

 a year every physicist of Europe knew of it and some had already de- 

 veloped new experiments on its basis.- Or consider Roentgen's paper 

 of 1896 ^ which might well be taken as the opening of the new physics. 

 The message which it contains was almost immediately broadcast all 

 over the civilized world; the necessary apparatus being available in 

 almost every physical laboratory, and the experiments being simple 

 enough they were promptly repeated in a hundred places; more than a 

 thousand books and papers on X-rays were published within the year 

 of their discovery.^ By the end of that year 1896, a physicist admitting 



^ Chronological diflBculties are not completely eliminated. For example, see 

 my paper "The discovery of conical refraction by William Rowan Hamilton and 

 Humphrey Lloyd in 1833" (Isis 17, 154-70, 1932). 



^ Facsimile reprints of the original Latin text and of the English translation ( Isis, 

 10, 435-43, 1928). 



^ The redaction of it was completed on Dec. 28, 1895, and it was immediately 

 printed, but it could hardly be distributed before 1896. See facsimile and Sarton's 

 analysis (Isis, 26, 349-69, 1937). E. Weil (Isis, 29, 362-65, 1938). 



* List of those 1044 books and papers in Otto Glasser: Roentgen (p. 422-79, 

 Springfield, Illinois, 1934; Isis, 22, 256-59). 



