RESPIRATORY METABOLISM 359 



measured respiration. Various workers have tried removing bacteria 

 by washing and filtering, or have corrected the figures for O, consump- 

 tion by running controls of bacteria without Protozoa. Data obtained 

 by these methods are extremely difficult for a reviewer to evaluate 

 critically, and usually one may either accept them at face value until 

 they can be checked with bacteria-free cultures or ignore them entirely. 

 In the present discussion the tendency has been to accept all data in 

 which the magnitude of the error is not obviously large, and to point 

 out possible difficulties involved. In view of the fact that some bacteria 

 have a respiratory rate per gram many times that of other types of cells 

 (the rate for Azotobacter is stated by Burk [1937] to be equivalent to 

 that of a 200-pound man consuming one ton of glucose per hour), the 

 present viewpoint may be considered far from conservative. In some 

 cases (e.g., the papers on cyanide insensitivity of Parmnec'nan) the data 

 on Protozoa seem quite adequate to prove the principal conclusions of 

 the author, but are not accurate enough to afford detailed comparisons 

 of respiratory rate. In such cases only the main points (e.g., insensitivity 

 to cyanide) are given serious consideration. In only a few cases have 

 investigators used bacteria-free cultures for measurement of respiration 

 (cf. Table 4). 



For comparative purposes in work with metazoan tissues, it is custom- 

 ary to express oxygen consumption in cubic millimeters (at normal tem- 

 perature and pressure denoted as N.T.P.) per hour per milligram of dry 

 weight of the tissue (symbolized by Qoa)- In the protozoan literature, 

 where the rate of O, consumption is expressed in absolute units, this unit 

 is sometimes the Q02 but is more usually mm^ per hour per organism, 

 principally because the counting of organisms is simpler than measuring 

 dry weight. Some authors use the symbol Q02 for Oo consumption per 

 1,000,000 or per 100,000,000 organisms (e.g., von Fenyvessy and 

 Reiner, 1928; Hall, 1938), but it seems preferable to avoid confusion 

 by retaining this symbol for its original meaning and using a new symbol 

 for consumption per 1,000,000 organisms, perhaps Q02 ^s used in Table 

 4. However, since not dry weight, nor wet weight, nor number of or- 

 ganisms affords the possibility of comparing the oxygen consumption per 

 unit of respiring protoplasm, these discrepancies are not as important as 

 one might at first suppose. In the case of flagellates such as Astasia and 

 Chilomonas , in which a high percentage of the weight may be in the 



