126 MASS. EXPERIMENT STATION BULLETIN 176. 



were supported in each one tlxrough notches cut in the cork stopper. The 

 different series were grouped as follows: — 



Plot 1. 

 Rye Seedlings. 

 Unlimed soil, unfiltered extract. 

 Unlimed soil, filtered extract. 

 Limed soil, unfiltered extract. 

 Limed soil, filtered extract. 



Clover Seedlings. 

 Unlimed soil, unfiltered extract. 

 Unlimed soil, filtered extract. 

 Limed soil, unfiltered extract. 

 Limed soil, filtered extract. 



The same arrangement was maintained for the soils of plots 6, 7 and 8, 

 and each extract was tested in three different bottles with a total of 12 

 seedlings. The cultures were maintained for four weeks, at the end of 

 which the seedhngs had begun to wilt. 



Differences in the seedUngs were noted by the end of the first week. 

 Those growing in the extracts from the limed soils were noticeably better 

 as a whole than those in extracts from unhmed soils. Rye seedlings in 

 the unhmed extracts had reddish stems and grew less rapidly. Roots of 

 the clover seedUngs in unlimed extracts began to appear stunted; es- 

 pecially so in the unlimed extracts from plots 6 and 8. When the experi- 

 ment was discontinued the best seedlings had developed in the extracts 

 from the limed soils of plots 6 and 8, while the poorest plants were in the 

 extracts from the unhmed soils of the same two plots. The roots of the 

 clover in these two extracts were short and thick and lacked branches. 

 Filtered extracts produced the same results as unfiltered ones. 



A lot of barley seedUngs was next used in the unfiltered soil extracts. 

 At the end of the first week the roots in the unlimed extract from plot 6 

 began to look stunted. By the end of two weeks the seedlings in aU the 

 unlimed extracts showed a tendency to wilt and the tips of the leaves 

 turned white. At the end of the fourth week, when the experiment was 

 stopped, the seedUngs in the extracts from the limed soils were uniformly 

 superior to those in the extracts from the unlimed. The poorest seedUngs 

 were in the extract from the unUmed soil of plot 6. 



The strikingly inferior growth of the different kinds of seedlings in the 

 extracts from the unlimed soils of plots 6 and 8, which had been dressed 

 with ammonium sulfate, suggested that the poisonous effect might be due 

 to sulfates of aluminium, iron or manganese, wliich were known to occur 

 in extracts from those soils. 



More culture experiments were accordingly tried from time to time, in 

 which standard nutrient solutions were used instead of soU extracts. Vari- 



