408 L. H. HYMAN. 



the diameters of the stems were determined under the compound 

 microscope with an ocular micrometer. However uniform a stem 

 may appear to the eye naturally under the microscope consider- 

 able variations in diameter are perceived. Some fifteen to thirty 

 measurements of the diameter of the coenosarc were made at 

 frequent intervals along each stem and these were averaged. The 

 length of the piece was measured on a millimeter rule. From 

 the average diameter and the length the volume of the piece 

 was calculated assuming it to be a cylinder. No correction was 

 made for the central cavity. The volumes of all of the apical 

 pieces in any one experiment were added together and the 

 volumes of the basal pieces similarly; and when the oxygen 

 consumed is divided by these volumes, the quotients can be used 

 to compare the rates of oxygen consumption of apical and basal 

 pieces. 



There is no doubt that considerable error is involved in such 

 determinations of the volume of ccenosarc in the pieces. It 

 does not seem to me, however, that the weight of the coenosarc 

 could be determined any more accurately. All determinations of 

 the rate of respiratory metabolism are necessarily erroneous 

 since there is no known way of discovering the actual quantity 

 of respiring protoplasm in an organism. The consistent results 

 which I have obtained in Tubularia, the definite relation noted 

 between level and oxygen consumption, and diameter and oxygen 

 consumption, indicate that the experiments are sufficiently 

 accurate to render the conclusions acceptable. 



Nine experiments were performed. The results are presented 

 in Table I. The first column of figures in this table gives the 

 oxygen content in cubic centimeters of the water in the tube 

 at the beginning of each experiment; the second column the 

 oxygen content in the tube at the end ; and in the third column 

 is given the difference between the first and second columns, 

 or the oxygen consumed by the pieces. The data are presented 

 in this way for the sake of simplicity; they are not actually 

 obtained in this form as the original oxygen content of the tubes 

 containing the pieces has to be calculated from the blanks. The 

 differences in oxygen content of the two tubes in each experiment 

 at the start are due simply to differences in the volumes of the 



