for Systematic Biology. 271 



used, a definitive natural classification, which we know to be 

 unattainable at the present stage of our knowledge. Hence as 

 long as this is the only available formula, the advice not to put 

 our very first attempts into final form must be endorsed by every 

 thinking man. This is just the point where my proposition comes 

 to the rescue. I shall try to show that it is possible to have 

 a formula which shall not mean definitive classification but shall 

 classify only so far as the facts allow and no further. With such 

 a formula we can both name, and at the same time lay a solid 

 foundation which time will complete into a natural classification, 

 no matter whether the groups are as stable as the Mammals or as 

 unstable as the stony Corals. 



Now, on the face of it, one would think that if only we know 

 what our aims are it cannot be impossible to begin our work with 

 those aims in view, in such a way that all good work should help 

 towards their attainment. What then is the ultimate aim of 

 classification ? Surely it is to arrange the organic kingdom in 

 the order of its evolution. This has long been accepted as the 

 ultimate aim of our attempts at classification, and the wiser of our 

 systematic workers acknowledge that their work is at its best but 

 a stumbling along in what appears to them to be the right 

 direction. But the question then is, Why do we continue to work 

 blindly, laying all sorts of fanciful foundations, which the next 

 worker roots up for another almost equally fanciful, burdening 

 himself at the same time by having to keep a faithful record of 

 all former attempts, however worthless they have been ? There 

 seems to me to be only one explanation, viz., that the old formula 

 of classification maintains too strong a mechanical hold over us, 

 and we have never seen our way to remodel it. This is hardly 

 to be wondered at when we think of the length of time the 

 Linnean method has been in use, of the indispensable services 

 it has rendered to science, and also of the fact that its in- 

 adequacy to meet the needs of the new evolutionary philosophy 

 is felt chiefly in work with very inconstant groups and, even when 

 felt most acutely, and understood most clearly, does not immedi- 

 ately suggest any remedy. It is further to be noted that the 

 whole formula as an ideal terminology for classification appears to 

 be quite adequate to the demands of the evolutionist, it being 

 doubtful whether we shall ever want other divisions than the 

 tribes, orders, families, genera, species and varieties. These 

 are all powerful reasons for retaining the Linnean formula for 

 naming the divisions of the organic kingdom. But that of course 

 is not the question under discussion, which is, How shall we 

 discover the divisions which we want to name ? It is quite 

 possible that when we have done the necessary preliminary sorting 

 out of the forms of life the divisions supplied by a Linnean 



