1922] ATWOOD—FORMALDEHY DE 251 
concern to avoid any fluctuations in respiratory results incident to 
the gaseous exchanges of saprophytes which might easily be con- 
fused with the results of seed treatment. NABOKICH, however, 
determined that the respiration of microorganisms on seeds may be 
disregarded during the first day, counting from the time of wetting 
the seeds. It is thus believed that the data given here eliminate 
the errors incident to such secondary factor. 
Throughout the work over forty runnings were made, represent- 
ing over 160 different seed lots. As regards the possible criticism 
that the several per cent variation in moisture content based on 
dry weight might make the results incomparable, it must be borne 
in mind that absolute carbon dioxide yields of different runnings 
are not to be compared with each other, but only the four lots 
used in any one run. Careful analysis of the variations in carbon 
dioxide output as related to moisture content has indicated that 
these variations may not be ascribed to moisture content differences, 
these observations being made in duplicate independently for the 
four seed lots of any one running. Fig. 10 and table II summarize 
the entire results of the investigations on respiration. In all of the 
work care was taken to have present in the flasks and bead towers at 
least twice as much of the alkali as would be neutralized by the 
carbon dioxide liberated by the seeds during any one run. The 
graphs are expressed in terms of the percentages of the barium 
hydroxide neutralized. In each case 25 cc. of fourth normal 
barium hydroxide was used, and at the close titrated against fourth 
normal hydrochloric acid in the presence of phenolphthalein. If, 
for instance, 12.5 cc. was neutralized the graph would express 50 
per cent values. 
Fig. 10 shows marked depression of the respiratory rate for the 
highest concentrations (1-80) as compared with the water soaked 
controls. The depression of the respiration rate is evident, although 
decreasingly so, at 1-160, 1-240, and down to 1-320, the concentra- 
tion usually used in seed treatment. At 1-400 and 1-1000 the 
difference between the controls and the treated samples was neither 
SO great nor so constant as to indicate any marked effects of the 
formaldehyde on metabolism. Special care was used in checking 
out the situation at the 1-320 concentration, at which point 
