The light-bottle samples were counted 

 for 3 to 10 minutes to a minimum of 1 , 080 total 

 counts with an error of about 5 percent. The 

 dark-bottle samples were similarly counted, 

 although a mininnum total count of 400 for a 

 standard error of 5 percent was considered 

 acceptable. Due to pressure of time, few sam- 

 ples were counted beyond 10 minutes, even 

 though the desirable minimum number of 

 counts may not have been obtained. 



Every third count was a C^'* standard, 

 which afforded about 25,000 counts per minute. 

 This standard was counted for 5 nninutes. Each 

 sample was corrected to the standard count 

 immediately preceding or following it. 



If the dark -bottle sample differed in 

 time of illumination to any great extent from 

 the light -bottle sample, both counts were cor- 

 rected to a standard time, usually 6 hours. 

 Otherwise the time of illumination for the light - 

 bottle samples was accepted as the time for both 

 light- and dark-bottle samples. 



The total count was divided by the number 

 of minutes to give counts per minute. From this 

 figure the background counts per minute was 

 subtracted. Background counts per minute was 

 determined by placing a clean filter in the count- 

 ing chamber and counting for a period of 1/2 to 

 1 hour. 



In cases where replicate samples were 

 illuminated for the same length of time, the 

 values obtained for each (in counts per minute 

 minus background) were averaged Lf the difference 

 between the lowest and highest figures obtained 

 was not greater than 25 percent of the highest 

 value. If the difference was greater the calcula- 

 tions for the replicates were carried out 

 separately. 



Calculation of Rate of Carbon Fixation 



It is assumed that carbon fixation which 

 did not result from photosynthesis occurred uni- 

 formly in both the light and dark bottles. Also, 

 that photosynthesis did not occur in the dark 

 bottles, since light was excluded. 



The quantity of carbon photosynthesized 

 (fixed) per hour per cubic meter of water was 

 calculated as follows: 



where: 



(L-D) = 



the net C counts per minute, i.e. 

 the light -bottle (L) counts minus 

 the dark-bottle (D) counts. 



3= total counts per minute from the 

 known quantity of C^^ originally 

 added to each sample. 



= number of hours of illumination. 



The multiplier 24, 545 

 1000 •90-12 

 44 



derived from 

 converts carbon 



Mg. C/hr. / m. = 



(L-D) • 24. 545 



dioxide per liter of sea water (90 mg. , 

 according to Sverdrup et al. 1942: 189) to 

 carbon per cubic meter (atomic weight 

 of carbon is 12; molecular weight of 

 carbon dioxide is 44). 



No correction was made for isotope 

 effect, nor was any attempt made to compute 

 more accurately the actual concentration of car- 

 bon dioxide in individual sea-water samples. 

 Such adjustments, while they might improve the 

 accuracy of the method, were considered for 

 the time to be meaningless refinements since it 

 was felt that they represent corrections that lie 

 within the errors introduced in the mcinipulations 

 jund calculations. The calculations were done 

 independently by two individuals and any differ- 

 ences rectified. The results, which we consider 

 as "raw data, " are listed in table 6. 



Chlorophyll Determinations 



Approximately 10 ml. of 90 -percent 

 acetone was added to each vial containing a 

 Millipore filter disc. The vial was then placed 

 in the dark in a refrigerator for 12 to 24 hours 

 to permit extraction of the pigments. After this 

 extraction period the contents of the vial were 

 decanted into a 15 -ml. centrifuge tube. The vial 

 was rinsed twice with 2 or 3 ml. of the acetone; 

 after each rinse the contents of the vial were 

 added to the centrifuge tube. The pigment ex- 

 tract was centrifuged in a clinical desk centri- 

 fuge at 3,000 to 4,000 r.p.m. for 5 minutes. 

 The extract was then deccinted into 5-cm. ab- 

 sorption cells. The residue in the centrifuge 

 tube was resuspended in 1 to 2 ml. of acetone. 

 After the residue had been allowed to stand for 

 2 or 3 minutes, it was again centrifuged at 

 3,000 to 4,000 r.p.m. for 5 minutes. The 

 extract from this residue was then added to the 

 absorption cell. The cell was filled with ace- 

 tone and inserted into a Beckman Model B 

 spectrophotometer. With the sensitivity control 

 set at the number 4 position, transmittance 



11 



r^ 



