PREFACE TO THE THIRD EDITION. {{\ 



In estimating the probable consequences for tlie long run, it is necessary to 

 discriminate between any given ornithological fact and the handle we may agree 

 to give that fact. The former is a natural fixity, the latter is a movable furni- 

 ture ; the former is subject to no autliority we can set up, the latter is wholly ar- 

 bitrary, determinable at our pleasure. Uniformity of nomenclature is so obvious 

 and decided a practical convenience that even at the risk of seeming to laud 

 work in which he had a hand, the author cannot too strongly urge compliance 

 with the Union's code, and adherence to the set of names the Union has 

 adopted. These may not be the best possible, but they are the best we have. 



The author's insistence upon this point does not of course extend to any 

 case where an error of ornithological fact may appear. That is an entirely 

 different matter. Reserving to himself, as he certainly does, the right of indi- 

 vidual judgment in every question of ornithological science, he is the last to 

 persuade others to refrain from equal freedom of expert opinion. " So many 

 men, so many minds," even when the number is only five ; no individual opinion 

 is necessarily reflected upon any point in the Code and Check-list ; it is the collec- 

 tive voice of a majority of the Committee that is heard in every instance. The 

 occasion for individual dissent on the part oi" any member of that body, as of any 

 other writer upon the subject, arises when in his private capacity as an author 

 he has, as it were, to pass upon and approve or disapprove any results of the 

 labors of others. The Appendix to the present edition of the " Key " unavoidably 

 brings up such an occasion. Yet that he may not even seem to reflect upon any 

 of his co-workers, his (-riticism express or implied has been sedulously reduced 

 to its lowest terms. It consists chiefly in declining to admit to the "Key" some 

 forms that the Committee have deemed worthy of recognition by name. Indeed 

 he has preferred to err, if at all, on the other side, desiring to give the user of this 

 book the later results of the whole Committee. 



Xevertheless he must here record an earnest protest, futile though it may 

 be, against the fatal facility with which the system of trinomials lends itself to 

 sad consequences in the hands of immature or inexperienced specialists. No 

 allusion is here intended to anything that has been done, but he must reiterate 

 what was said before ( Key, p. xxvii ) respecting what may be done hereafter if 

 more judicious conservatism than we have enjoyeil of late be not brought to bear 

 down hard upon trifling incompetents. The " trinomial tool " is too sharp to 

 be made a toy ; and even if we do not cut our own fingers with it, we are likely 

 to cut the throat of the whole system of naming we have reared with such 



