i6 
changed the names employed by their predecessors, the intention 
has rarely, if ever, been that of piracy, but rather a conscientious, even 
if ill-judged, wish to improve upon faulty or inappropriate designa- 
tions given by earlier writers. (2) The description of a supposed 
new species is by no means a difficult matter, but the proper generic 
affinities of the plant in question are often to be determined only by 
the most critical and discriminating investigation. It thus happens 
in many cases that the service of the original describer, far from 
being greater, is decidedly less, than that of the author of the first 
correct combination. At all events, it would be unwise to sacrifice 
to a sense of justice so sentimental any provision which is likely to 
conduce to stability and simplicity. (3) In a comparatively small 
number of cases a diverse interpretation of generic names and limits 
would doubtless lead to the simultaneous use of different specific 
names for the same plant. It is believed, however, that this danger 
has been considerably over-estimated by those who have advanced 
the objection. Maia 7 ithemiim Corivallaria and its synonyms have 
been several times used as a drastic example of the baleful effects of 
the rule under consideration, but a very obvious answer to objections 
of this kind, is that under a properly drawn code it should be impos- 
sible for four or five different generic names to be simultaneously 
current for the same genus, and that, as the possibilities of the pres- 
ent widely different use of generic names are diminished by the 
more detailed codification of nomenclature, any annoyance from dif- 
fering specific names for the same plant under different genera will 
of necessity be much decreased. (4) As already mentioned the his- 
torical development of classificatory nomenclature in phanerogamic 
botany has differed considerably from that of cryptogamie botany ; 
and both have long had a wholly independent development from the 
nomenclature of zoology, habits convenient to the particular subject 
having been formed by investigators in each of these fields. Under 
these circumstances it seems highly inexpedient to make in any one 
of these subjects any important sacrifice to secure a superficial and 
relatively unimportant agreement of method. Surely any investi- 
gator with a capacity so great as to work successfully in more than 
one of these wide disciplines, should be able to grasp and apply 
without particular difficulty two or three slightly different systems of 
nomenclature. 
The chief reasons for maintaining the first binomial in dealing with 
