128 



BACTERIA IN RELATION TO PLANT DISEASES. 



The effect of stunting in preventing general infection is shown quite clearly in Series 

 III to VII (1902) and in the field experiments at Arlington in 1903. It is also shown in 

 the following experiment. 



SERIES XX TO XXIV, 1908. 



This experiment was begun July 14-15, 1908, on another part of the grounds of the 

 U. S. Department of Agriculture, out of doors, in a vacant cold frame between two hot- 

 houses. The same Golden Bantam sweet corn was planted as in Series XVI to XIX, but 

 the soil was different, and the other conditions were such that it was known from the start 

 that the plants would be badly dwarfed. In fact, the experiment was made for this very 

 purpose, so that the effect of slow growth might be compared with more rapid growth such 

 as was anticipated and obtained to some extent in the preceding experiment. The corn 

 was planted rather thickly in new 6-inch pots. Good hot-house soil was used. This was 

 prepared from well-rotted sods thoroughly mixed with old cow manure which had been on 

 the Department grounds in compost heaps for the previous 9 months. The cold frame 

 was divided into five compartments and planned as follows : 



(1) Two hundred and seventy-five pots, planted with selected good seed, about 6 seeds per pot, 

 more rather than less, soaked 8 minutes in i : 1000 mercuric chloride water, and then washed in sev- 

 eral waters. On July 21, 62 1 plants were visible. About one-half the seeds germinated. 



(2) Two hundred and seventy-five pots, planted with selected good seed, about 6 seeds per pot, 

 more rather than less, soaked for 30 minutes in i : 200 formalin water. The formalin delayed the 

 germination. On July 21, 481 plants were visible. About one-third of the seeds germinated. 



The above two plots were planted on July 14 and the following next day. 



(3) Two hundred and seventy-five pots, planted with selected good seed, about 6 seeds per pot, 

 more rather than less, soaked for i hour in i : 2 hydrogen peroxide water. On July 21, 603 plants 

 were visible. About one-half germinated. 



(4) Six hundred pots, planted with unselected, untreated seed, about 8 seeds per pot, more 

 rather than less. On July 21, 875 plants were visible. Less than one-fourth germinated. 



(5) Three hundred and eighty-five pots, planted with selected bad seed, about 1 2 seeds per pot. 

 On July 21, 502 plants were visible. An average of about i in 6 finally germinated. 



The pots stood close together on clean sand. The average height of the plants when 

 harvested was between 2 and 3 feet, i. e., they were much stunted by lack of room for the 

 roots. Many of the seedlings showed, soon after germinating, what were regarded at first 

 as suspicious signs, and plants in 74 of the pots were marked, but nothing developed later. 

 The suspicious signs were water-soaked streaks on the young leaves. These turned out 

 afterwards to be due to excessive pumping out of water through the water pores and were 

 not infections. Some of the leaves recovered promptly; others retained yellowish stripes 

 fora long time, but none contracted Stewart's disease. On September i (forty-seventh day) 

 a careful examination of all the plants showed no cases. The corn had begun to tassel. 

 It was at that time also badly dwarfed and yellowish from lack of room for the roots. The 

 tallest and greenest plants were next to the hot-house, where they were somewhat shaded 

 from the morning sun, and did not, therefore, dry out as quickly. The first cases were 

 observed at the end of about 60 days. The plants were harvested September 29. The 

 per cent of cases by plots is shown in table 16. 



TABLE 16. Efect of Slow Growth on Development of the Disease. 



