30 



field experiments the use of bur clover earth has pro- 

 duced tubercles and greatly increased the yield of al- 

 falfa. 



Figure 2 shows typical alfalfa plants taken in April, 

 1903, from three plots at Auburn, sown the preceding 

 October. The small plants on the right liad been neither 

 limed nor inoculated and were free from tubercles ; those 

 in the center had been inoculated, but not limed; the 

 largest plant had been inoculated and limed, and here 

 the supply of tubercles is abundant. Soil from an old 

 alfalfa field was used in this experi'ment as inoculating 

 material. 



INOCULATION OF AI>FALFA ON PRAIRIE SOILS. 



In the light of Dr. Hopkins' demonstration we 

 can now see why it is unnecessary to inoculate 

 alfalfa on fields where melilotus (sweet clover) 

 has recently gTOwn, and produced tubercles, as 

 it almost invariably does on prairie soil. It 

 would still seem advisable, however, to inoculate 

 alfalfa seed to be sown on such prairie land as has 

 not recently grown melilotus. While these germs have 

 probably l)een widely distributed by wind and water and 

 otherwise, in the prairie region, we have no proof that 

 they are present in all fields of prairie land in sufficient 

 numbers for best immediate results with alfalfa. 



Indeed the observation made by Mr. J. O. Hays in 

 our fertilizer experiment on gray prairie land in Greene 

 county, pre^doush' referred to, seems to indicate that 

 there is an advantage, at least during the first few 

 mouths of growth, in inoculating alfalfa on some lime 

 land. In 1902, on land which had been used as a pas- 

 ture for a number of years, he reports that on the six- 

 months-old inoculated alfalfa plants tubercles were 



