1910] DACHNOWSKI—ARID HABITATS 331 
cultures. Several flasks were left sterile to serve as controls, while 
others were inoculated with a mixture of the various species and with 
the bacteria found in 1°¢ of fresh bog water. An additional test 
condition was arranged at the same time from the normal untreated 
bog water. All flasks were placed in an incubator for a period vary- 
ing from two to four or six weeks, and at such times were then 
brought to the laboratory. All physiological tests were made in dupli- 
cate series, and the greatest caution was observed to reduce the dangers 
of contamination during the preparation of the cultures. Wheat 
seedlings germinated in sterilized quartz and carefully washed in dis- 
tilled water were used as indicators. They were transplanted to 
half-liter “ Mason” jars in a manner as described in previous papers. 
From the data at hand the following have been selected to illus- 
trate the variation in virulency of bacterial products. Using the 
transpiration of the controls as a basis, and representing it as 100, the 
different bacterial cultures have values in the order as indicated in 
the last two columns of the tables below. 
These figures show conclusively that in all cases the bacteria are 
Tesponsible for the proportionally diminished transpiration and 
growth of the wheat plants. In all cultures the transpiration values 
lie below that of the control, varying from 20 to 52 per cent. The 
evidence derived from the duplicate series is omitted in tables II 
and IIT, showing as it does results closely parallel. 
More recent experiments with pure and mixed cultures of bog 
bacteria confirm the earlier experiments as to the ability of these 
organisms to increase the amount of toxins present in the peat sub- 
stratum. These results are significant in that they demonstrate the 
active participation of microorganisms in bog habitats, a fact con- 
trary to the position taken by Fritu and ScHROTER, who regard the 
bacterial influence as very small. To what extent the tables suggest 
the possibility that bacterial diagnosis when correlated with physio- 
logical criteria may determine the crop-producing power of different 
agricultural soils need not be discussed here. 
Chemically peat is made up of substances which are commonly 
classified among the dehydration products of the carbohydrates. 
The analyses which have been made agree in showing that the phos- 
phoric acid content of peat soils is high and partially in an available 
