RESEARCH COLLECTIONS 183 



only be exceedingly numerous (so numerous, in- 

 deed, that it is almost impossible to put a limit to 

 what may be required), but they must also be kept 

 under such conditions as to admit of ready and 

 close examination and comparison. . . . 



'' Collections of this kind must, in fact, be treated 

 as books in a library, and be used only for con- 

 sultation and reference by those who are able to 

 read and appreciate their contents. To demand, 

 as has been ignorantly done, that all the specimens 

 belonging to our National Museums, for instance, 

 should be displayed in cases in the public galleries, 

 would be equivalent to asking that every book in 

 a library, instead of being shut up and arranged on 

 shelves for consultation when required, should have 

 every page framed and glazed and hung on the 

 walls. . . . 



"In the arrangement of collections designed for 

 research, the preservation of the objects from dust, 

 light, and damp ; their absolutely correct identifica- 

 tion and record of every circumstance which can 

 be known of their history ; their classification and 

 storage in such a manner that each one can be 

 found without difficulty or loss of time, and that 

 they should be made to occupy as small a space as is 

 compatible with their requirements, both on account 

 of expense as well as of convenience of access, 

 must all be principal aims. They should be kept 

 in rooms provided with suitable tables, and a good 

 light for their examination, and within reach of the 



