182 



BACTERIA IN RELATION TO PLANT DISEASES. 



etc., he ma}- be trusted to undertake some original research. Even when once on 

 his feet as an investigator, my advice to him would be : Try every conclusion 

 repeatedly and make haste slowly. When he becomes uneasy at delays, let him 

 reflect that one really good paper does much to set an unknown worker on his feet 

 among scientific men, whereas one or two hastily written, poor papers will injure 

 his reputation as an investigator more than half a dozen good papers subsequently 

 published will suffice to repair. Moreover, in this age of enormously multiplied 

 publication it is impossible to read everything, and consequently if a writer wishes 

 to attract attention he must have a commanding grasp of his subject ; must present 



Fig. 144.* 



its leading features in a clear, interesting style; must be as brief as the importance 

 of his subject will warrant, otherwise his words are certain to be overwhelmed and 

 lost; and, finally, must publish in a proper place, /. c., not in some obscure "Transac- 

 tions" or in some local journal with a small circulation. II 'hen ready to publish, 

 stop and do your work all aver again it'it/i more care. This is my advice to begin- 

 ners. In the course of such general revision the chances are that many statements 

 will require correction or modification, and some may have to be omitted altogether. 



*FiG. 144. Colonies of Bacillus aroideae, circular when grown on an agar plate at 37 to 38 C., 

 i. c., at ;\ temperature near the maximum. Photograph hy Townseml. 



