No. 4.] GERMAN EXPERIMENT STATIONS. 397 



to control the many conditions sufficiently to draw accurate 

 conclusions from the results obtained. 



6. When one series of experiments is completed, it will 

 require years to bring the plats into condition suitable for 

 another series. 



7. It is impossible to establish the laws of plant nourish- 

 ment by means of results obtained from field experiments. 



The field experiment has its place, however. It should give 

 us information concerning local conditions, concerning the 

 characteristics of various soils which ought to be ascertained, 

 in order to get the most value from what science has taught. 

 On the other hand, it is the duty of the investigator to study 

 the workings of tlie different fertilizers, ivJten the experiments 

 are entirely beyond the control of all local influences and 

 conditions. 



Believing, therefore, that results could be more quickly and 

 more accurately obtained in another way, this investigator, 

 some eiiiht years aii'o, turned his attention to the so-called 

 pot experiment. He now uses pots made of tin, Avhich are 

 placed upon })latform cars. These cars set upon iron tracks, 

 and can be run under cover in case of a severe storm. The 

 pots vary in size from six to twelve inches in diameter, and 

 from nine to fifteen inches in depth. They are filled with 

 about an inch of gravel, and then with the thoroughly mixed 

 soil. A small tulje extends down along the side to the bot- 

 tom of the pot, to conduct the water below, instead of 

 })ouring it on the surface. The latter method of watering 

 renders the soil too com})act. The fertilizers are accurately 

 weighed out in the laboratory, and carefully ai)plied by 

 either the professor or by a trusted assistant. Large iron 

 cylinders, var\-ing in size from a keg to a barrel, open at 

 both ends, are sunk in the ground and filled with thor- 

 oughly mixed soil, and are used for large growing plants. 



The soil is previously prejmred by subjecting it either to 

 a general or partial exhaustion. 



When the writer was at the Darmstadt station, a year ago, 

 Professor Wagner had some 1,100 to 1,200 pots in o[)era- 

 tion. The professor states that by this method he can make 

 a very much larger numljer of experiments than by the or- 



