92 



SCIENCE. 



[iM. S. Vol. XXIII. No. 577. 



able measure of uniformity and which 

 goes far to establish the rank of such cate- 

 gories as genus, species and variety. This 

 fact is clearly shown by the contrasting 

 work of those free-lances who, armed with 

 the less effective weapons of a more re- 

 stricted knowledge, have in doing inde- 

 pendent battle upon the difficulties of gen- 

 eric classification followed other tactics 

 and set up new standards. I doubt if they 

 have realized how quickly and fully the 

 personal equation is recognized in regard 

 to their work, or how generally even the 

 amateur and layman grasp the fact that 

 their generic and specific propositions are 

 not up to the standard. No one can change 

 the temperature by making the degrees of 

 his thermometer smaller. Least of all is 

 it possible to make people believe that the 

 shortened degree is as important as the 

 longer one. Time spent in this mere let- 

 ting down of standards and shifting of 

 ranks is worse than wasted. The process 

 is annoying and confusing, for to the nat- 

 ural difficulties of generic classification 

 plus a certain inevitable historic element 

 of artificiality, it superimposes the most 

 awkward and irritating difficulty of all, 

 namely, the personal equation. 



Let us get something done and not spend 

 our time in endless and profitless strife 

 about first principles, thereby bringing 

 confusion into what may be regarded as 

 fairly well established already. There are 

 limitless fields for further profitable work 

 in the finer classification of the flowering 

 plants without perpetual tampering with the 

 boundaries of important and long-studied 

 genera— a type of activity very prone to 

 sink to the level of a mere juggling with 

 names. Having said so much against gen- 

 eric changes of a superficial character, I 

 fear some of my hearers may get the im- 

 pression that I- am opposed to generic 

 changes in general and perhaps even to 

 the further inve.stigation of generic limits; 



but this is in no wise the case. There is 

 certainly great opportunity for further and 

 very profitable study of generic classifica- 

 tion. The genera of several families, as 

 for instance the Cruciferffi, are in many 

 cases pretty artificial groups. We need 

 much further knowledge of the relation- 

 ships of the species concerned. Let those 

 who wish to be of real service in this mat- 

 ter give us what we so much desire, namely, 

 additional light upon the ontogeny, embry- 

 ology or finer anatomy of these species, 

 sources of information sure to yield data 

 of high classificatory importance. 



In closing let me urge that, while we 

 remit no effort to secure further light on 

 this subject, there should be a general 

 agreement to treat the accepted and tradi- 

 tional interpretation of large and impor- 

 tant genera as sacred and binding imtil we 

 can furnish definite and convincing evi- 

 dence that change is needful, and that for 

 the welfare and dignity of our science, all 

 should unite in opposing changes of the 

 artificial sort, which consist merely in the 

 shifting of ranks and modification of stand- 

 ards. 



B. L. Robinson. 



Hakvaed University. 



INVESTIGATIONS AND COMMERCIAL TESTS 



IN CONNECTION WITB THE WORK OF 



AN ENGINEERING COLLEGE.' 



In any school it is necessary, in securing 

 the best efficiency in instruction, that the 

 professors shall be able to speak with au- 

 thority on the subjects which they teach. 

 In technical schools those who teach the 

 practical engineering subjects can not speak 

 with authority unless they have had prac- 

 tical experience. Investigations and com- 

 mercial tests may serve to give them this 

 practical experience, and the question nat- 

 urally arises— is it a good policy for pro- 



' Address of the vice-president and chairman of 

 Section D — Mechanical Science and Engineering, 

 at New Orleans, December 29, 190.5. 



