400 BOTANICAL GAZETTE [MAY 
The seeds tested were suspended from platinum baskets at A, 
being laid on moistened neutral asbestos fiber, the non-organic 
substance being used to prevent drying out during the test, and 
at the same time to avoid danger of respiratory changes. Under 
the baskets were placed small vessels, each containing 1 cc. of 4 
per cent potassium hydroxide solution to absorb any carbon dioxide 
evolved during the period. Test tube D was filled with mercury 
and could be raised and lowered to balance the mercury column 
in C with that in the test tube. 
Tube C was 15 cm. long and hada 
volume of 2 cc., the graduations for 
which were made so that readings 
could be made to o.or cc. When 
the rate of oxygen intake from the 
air was to be measured, the gas in 
the apparatus was previously dis- 
placed for several minutes with a 
current of air forced through 40 per 
cent potassium hydroxide solution , 
to remove any traces of carbon 
dioxide. Other gases tested were 
passed through the alkali and 
apparatus in like manner. During 
the course of taking the readings, 
the apparatus was submerged in a 
Freas constant temperature water 
bath electrically regulated and accu- 
rate to about o°01 C. The volume 
of the chambers was so selected that variations in temperature 
would introduce an error so small as to be negligible with the water 
bath used. Thus, for the right chamber a variation of 0°! c. 
would result in an error of 0.0056 cc., and for the left chamber of 
0.0062 cc.. Corrections for barometric and temperature variations 
were made; hence any rise of the mercury column as shown by 
the corrected readings indicated absolute absorption of oxygen by 
the seeds tested. From the basis of the dry weight of the seeds 
used the rate of oxygen absorption per gram per hour was com- 
Fic. 7.—Respirometer 
