No. 6. DEPARTMENT OP AGRICULTURE. 315 



9. Land, Newark, Del., under active cultivation aur- 

 ing sumaier, now in crimson clover (Chester, 

 1!)01),« 1,880,000 



10. Soil from the center of a strip of woodland, New- 



ark, Del. (Chester, 1901),^ 70,000 



11. A family vegetable garden, Newark, Del. Rich 



in humus and actively cultivated,*^ 1,81G,000 



The preceding table, except No. 10, represents agricultural soils. 

 In special instances the number may rise much higher, particularly 

 in soils in the immediate vicinity of dwellings and stables. Accord- 

 ing to Manfredi,2i the number of bacteria in the dust of the streets 

 oi Naples varied from 1,000,000 to 10,000,000 per gram, and even 

 higher. Maggiora^-^ gives figures as high as 78,000,000 for the num- 

 ber of bacteria per gram of soil in certain inhabited spots. 



Adametz in 1886,^ Fraenkel^ in 1887, and Caron^ in 1895, showed 

 that the maximum number of bacteria were not found at the surface 

 but at a depth of from nine to eighteen inches beneath the same. 



In 1887 Fraenkel^ showed that at a depth of from thirty to sixty 

 inches there was a rapid and abrupt diminution of the number of 

 germs from 200,000 at twenty inches, to 2,000 at thirty-nine inches, 

 while at a depth of five feet no living germs were obtained. These 

 results are not altogether in accord with results obtained by the 

 writer in Delaware, which show that the maximum number of bac- 

 teria occurs in the first six inches of soil, below which they diminish 

 at a very rapid rate, until at twenty-four inches only about one-five- 

 hundredth of the number at the surface exist. Furthermore, it was 

 found that the highest numbers exist not at the surface, but at a 

 depth of about four inches below the same. In the following figure 

 2 is shown the rapid decline in the number of bacteria in the soil as 

 the depth increases, determined at the Delaware station. 



VI. CONDITIONS AFFECTING THE NUMBER OF BACTERIA 



IN THE SOIL. 



The observations of Maggiora in 1887 have shown (1) that the num- 

 ber of germs in desert and forest soil is much smaller, other things 

 being equal, than jn cultivated lands; (2) that the number is propor- 

 tionate to the activity of cultivation and the strength of fertilizers 

 used, and (.3) that light sandy soils contain fewer germs than those 

 rich in clay and especially those rich in humus. 



These results are in accord with those given in the preceding 

 table, which show (1) the very low number present in woodland 

 soil, (2) the very high number present in soils under active cultivation 



