4. How many abstracts have proved useful in 

 locating information? 



some useful to ;>9.3 per cent, of readers, 

 few useful to 16.6 per cent, of readers, 

 none useful to 11.4 per cent, of readers. 



87.3 per cent. 



5. Should the abstracts be continued? 



yes 92.8 per cent, of readers. 



HO 4.0 per cent, of readers. 



96.8 per cent. 



Since the abstracts are read, always or usu- 

 ally, by 88 pev cent, of the readers; are read 

 instead of many of the articles by 45 per cent, ; 

 and have been- found helpful in other ways by 

 59 per cent., they have evidently proved of de- 

 cided value. In what ways? According to the 

 testimony of the readers: "They are great time 

 savers"; "They frequently give all information 

 necessary about the articles"; "They double a 

 man's range of reading." Moreover, the ab- 

 stracts have 'been copied verbatim in Science 

 Abstracts and have rhus reduced the labor of 

 preparing that abstract journal by ten per 

 cent. The abstracts have therefore to seme 

 extent fulfilled their purpose of saving the 

 time of scientific men. 



How about other sciences? Are astronomy 

 and physics essentially different from chem- 

 istry, botany, zoology, geology, physiology, 

 etc., in their methods of disseminating scientific 

 information? If 93 per cent, of the readers 

 of these as.tronom.ical and physical journals 

 find that preliminary abstracts are useful to 

 them, would not the great majority of other 

 scientific readers also find such abstracts use- 

 ful? The responsibility resting upon the 

 editors of other scientific journals is clear. 



Now as to the nature of the abstracts which 

 should be provided. The abstracts which have 



