394 



THE AMERICAN NATURALIST [Vol. XLIV 



ate the present uncertainty and confusion, and would 

 open the door to those who, regarding nomenclature as 

 an end in itself and not merely a means by which the 

 necessary evil of naming plants can be reduced to a min- 

 imum, devote time and labor to the undesirable task of 

 unearthing names which are at the best uncertain, at the 

 sacrifice of names which have been in universal use for 

 many years, and whose meaning is perfectly clear. To 

 my mind the object should be, not to attempt to seek per- 

 fection in authority and priority— a hopeless task— but 

 rather to select the best solid basis in some comprehen- 

 sive work. Even then, there is the question of genera 

 conservanda and I believe that, whatever work or date 

 is adopted, it will be most desirable to adopt a list of 

 genera conservanda. There is nothing illogical in this 

 and practically there are great advantages unless one be- 

 lieves in the theory that mere changing of names is a 

 merit in science. That theory I certainly do not accept, 

 but hold that the fewer changes of names the better. 



It has been my misfortune never to have found any- 

 thing perfect. Some of my friends have perfect systems 

 of classifications of books, of herbaria or of plants. In 

 trying to apply perfect methods I always recall a visit 

 in company with Sir Joseph Hooker to an establishment 

 not a thousand miles from here. The person in charge 

 said, "We think we have a perfect museum case which 

 we would like to show you. " ' * Yes, ' ' said Sir Joseph, 1 ' I 

 am always glad to see what I have never seen. But what 

 do you keep in the case!" A key was produced, but by 

 no amount of coaxing or forcing could the case be opened. 

 ' 'Yes," said Hooker, "I presume that it is perfect, but I 

 prefer cases which open.'' The same remark would 

 apply to a good many systems. They are perfect until 

 we try to find out what is in them. 



