Encyclopaedias 



83 



opinion of earlier times. Unfortunately, the old encyclopaedias are difficult to con- 

 sult, because even when they are available as they are in the larger libraries, they 

 are generally hidden away on the theory that they are obsolete and superseded and 

 that nobody will ever want to consult them.** That practice is certainly wrong 

 as far as the historian of science is concerned. Indeed, encyclopaedias are not avail- 

 able except when they are completely available on open shelves. When the historian 

 wishes to consult them to investigate the evolution of ideas (say, on the speed of 

 light), he will generally wish to consult not one of them but a whole series, and in 

 many cases he will not know which particular volume to ask for (the information 

 ad hoc might be given under fight, or optics, or speed of light, or even elsewhere). 

 It would be impracticable to borrow every one of those bulky series, each time that 

 a similar investigation had to be made. 



An Institute for the history of science should include an "encyclopaedia room" 

 where all the new as well as the old encyclopaedias could be easily consulted. For 

 example, there ought to be a full set of all the Britannicas. The same room might 

 contain also ( if space permitted ) other reference books such as the biographical col- 

 lections (to be described presently), gazetteers, dictionaries and grammars. 



** Many of the old encyclopaedias owned by the Harvard Library are stored away in the Deposit 

 Library across the river, and cannot be consulted except after their return from Deposit to 

 Widener; this may take a few days. 



