PERSONAL HABITS 



this rapidity of action never led to the slightest dis- 

 courtesy, nor to the neglect of anything essential. His 

 manuscripts for the printer bore the marks of incessant 

 corrections, and he never hesitated to alter and cut at 

 pleasure until the word to print was finally given. He 

 has been heard to say, " I cannot tell how a paragraph 

 will look until I see it in type." 



Dana's study was in his dwelling-house. It was a 

 bright, sunny room facing to the southwest, with a large 

 anteroom which served as " a stack " for such books as 

 were not in frequent use. His working apparatus was 

 simple a few instruments, a small cabinet, a good many 

 maps, and a library of moderate size, chiefly composed 

 of scientific works. There was a side door to the north 

 by which the family maintained easy access to their kin- 

 dred next door an entrance, moreover, by which many 

 of those who were accustomed to consult the editor 

 in his sanctum had the freest admission. They would 

 appear without being announced, and their host, when 

 he was well, would readily lay down his pen and engage 

 in conversation; or, more frequently, he would proceed 

 to the correction of a proof-sheet, or the preparation of a 

 note, or the draughting of a letter on some subject intro- 

 duced by his visitors. He had the art of bearing inter- 

 ruptions gracefully and of turning again to his work as if 

 nothing had occurred. It was his custom to be his own 

 letter-carrier, and two or three times a day he might be 

 seen going to and from the post-office, hands, pockets, 

 and even hat filled with the voluminous mail that per- 

 tained to the Journal of Science. His library was a 

 laboratory. It overlooked the garden where he often 

 spent an hour of repose in the care of his plants and 

 shrubs. He was not a buyer of many books, but every- 

 thing in his line seemed naturally to seek him. The 

 shelves were filled with the transactions of the learned 

 societies to which he belonged, long sets of scientific 



281 



