12 



BACTERIA IN RELATION TO PLANT DISEASES. 



tissues, it is sometimes desirable to grow them for a day in bouillon before attempt- 

 ing the plate-cultures ; but one must then be on his guard, since it is quite possible 

 by this method to start with enormous numbers of the right organism and have the 

 bouillon culture filled with something else at the end of the 24 hours. 



Pure cultures may also sometimes be obtained by cutting out pieces of the 

 tissue and throwing them into tubes of culture media. This method, however, 

 shows little or nothing as to the prevalence of the organism in the tissues, and in 



$x 



-'^'W"^''" 



-mi 



.-^IfSSrr 



pssi 



'Av'. ?1r. ! ^ -. _-'v- J.l 



^j|* ( ||-, 



^%K : ||^B 



7^5^^; 





/ '"^^^ 



WMK 



Fig. 7* 



the hands of beginners is very liable to miscarry. If growth is obtained it may 

 indeed have come from man}' organisms of one sort pervading the tissues and 

 causing the disease, but it is not certain that it did not result entirely from one or 



*Fic. /. Bundle in a cauliflower-petiole entirely destroyed by Bacterium camfestrc. The re- 

 sult of a pure-culture inoculation. Plant No. 112 inoculated March 10, 1897, by needle-punctures on 

 the blade of a leaf .without 'hypodermic injection. First signs of disease March 20. Petiole put into 

 alcohol on April 5. Longitudinal section. Tissues surrounding .the bundle entirely free from bac- 

 teria. Section not made from the inoculated leaf, but from the first leaf that showed secondary 

 signs. Drawn from photomicrograph of a paraffin section stained with carbol-fuchsin. X 206. 



