2. BULLETIN 245, U. S. DEPARTMENT OF AGRICULTURE. 
Arlington were carried out with emulsions of this substance and fish- 
oil soap. The strength of these emulsions and the results obtained 
are given in Table II, series 95 and 101. The larvicidal results, with 
the exception of series 95, C, were good. The emulsions apparently 
produced a considerable increase in numbers of the bacteria. An 
increase of water-soluble nitrogen and ammonia was obtained in all 
the treated samples. Some further experiments at New Orleans, 
using the cages which were employed in 1913, gave poor results, as 
many larvee escaped (Table III, series 109 and 113). The results of 
six open-pile experiments are recorded in Table V(series 42, A, B, C, 
and D, and 52, A and B), and the data for two additional open-pile 
experiments are given in Table IV (series 47, A and B). Bacterio- 
logical and chemical results are given in connection with the last two 
experiments. The duplicate samples indicate that there was a slight 
reduction in the number of bacteria and that there was no apparent 
effect on the manure as shown by the chemical data. All the exper- 
iments indicate satisfactory larvicidal results, but the best were those 
obtained with the largest quantity of fish-oil soap, namely, 1 pound 
(Table V, series 52, A and B), which killed 93 per cent of the larve. 
This fact suggests that the fish-oil soap is an important constituent 
of this larvicidal mixture. 
OXALIC ACID. 
Four cage experiments were carried out with oxalic acid, using 1 
and 2 pounds to 10 gallons of water. One experiment with 1 pound 
gave negative larvicidal results, while the results from three experi- 
ments using 2 pounds were as foliows: 3 per cent, 60 per cent, and 80 
per cent. In the one sample of manure analyzed the bacterial count 
was reduced, and the ammonia was increased over the control. 
PYRIDINE. 
Pyridine (C;H,N), which is prepared commercially from coal tar, 
and is also obtained from the distillation of bone oil, is alkaline, con- 
tains 17.75 per cent nitrogen, and costs about $1 per pound. This 
liquid was used in three cage experiments in dilutions of 1 to 100 and 
in one cage experiment in a dilution of 1 to 500. The results of the 
cage experiments are shown in Table I (Series 78 and 104). The 
larvicidal efficiency of the 1 to 100 dilutions was 63 per cent for one 
cage and 99 per cent for the other two. The 1 to 500 dilutions showed 
no apparent effect. No consistent action on the bacteria is evident, 
and the water-soluble nitrogen and the ammonia from the treated 
samples are higher than from the controls. 
In open-pile experiments at New Orleans (Table V, Series 41, A 
and B) pyridine 1 to 500 was used twice, giving an apparent larvicidal 
