8 
which are available in the class of material that must 
be studied. Perhaps the progress of American sys- 
tematic botany has been retarded more by the easily 
acquired habit of naming new specimens by matching 
them in the herbarium than in any other way; and the 
cing the limitation of species. Hence a specimen is 
always potentially better than a dese ription, since the 
former is the thing itself, while the latter is only a 
medical botany, as in general botany, the herbarium 
is of prime importance. 
he medical or pharmaceutic herbarium should 
differ from the general herbarium in containing only 
a 
of a million s The specimens contained in 
this eohioal herbarivim should be perfect from all 
points of view:—goo the sense of the ordinary 
botanist as representing the technical characters which 
rbarium 
needed by freshly gathered ea but dit ato 
