258 HITCHCOCK: STUDY of the local flora 



is at this point that the danger from crystalization occurs — ■ 

 the danger that the mobile mental attitude will harden. After 

 publication the worker may feel called upon to defend his state- 

 ments against the attack of critics. What before was a 

 detached investigation of facts becomes an effort at personal 

 vindication. All these troubles might be avoided by deferring 

 publication. Indefinite delay, however, is not advisable. Actu- 

 ally to finish an investigation before results are published 

 would usually mean no publication. The author should be 

 unhampered but should choose the happy mean between " rush- 

 ing into print" with half-prepared preliminary notes, and in- 

 definitely delaying publication till a perfect monograph may be 

 produced. It is of course desirable to ascertain before publi- 

 cation all the facts bearing upon a subject but sometimes our 

 hands are forced in such matters. Publication on a given sub- 

 ject may call for a statement concerning something which has 

 not been thoroughly investigated. All of which emphasizes two 

 points already mentioned, that equal weight should not be 

 given to all the statements of an author, and that the author 

 himself should correct his own errors as freely as he corrects 

 those of others. 



Our worker with Aster will have the opportunity for training 

 in another field, that of taxonomic judgment. It is well that 

 every scientist should have training in the two methods of 

 establishing facts, by experiment and by repeated observations. 

 Most facts in physical science are established by experiment; 

 most facts in descriptive biology are established by repeated 

 observations. In practice a taxonomist is called upon to clas- 

 sify individuals into species, and species into genera. A species 

 is not a law or a fact that can be proved by experiment. It 

 is a taxonomic idea which can be established only by a great 

 repetition of observations. Our classification is based upon the 

 evolutionary hypothesis that all living organisms are descended 

 from other somewhat different organisms of the past. Setting 

 aside the differences of opinion as to the independent origin of 

 the larger groups, it follows that all organisms, at least of family 

 groups, are probably genetically connected. Their lines of 



