SECT. 2] 



PRODUCTIVITY, DEFINITION AND MEASUREMENT 



149 



only to multiply by a factor of 2. In the tropics the effect of very high light 

 intensities is particularly conspicuous, for example the afternoon depression 

 such as that demonstrated by Doty and Oguri (1957). In this case it is important 

 always to use the same period for making experiments. 



The in situ method is time-consuming even when the experiments last only 

 from noon to sunset. It is, therefore, in most cases too expensive to use on 



10 20 



mg C/m day 



10 20 30 



mg C/m day 



10 20 



mg C/m day 



Fig. 9. In situ experiments in the Kattegat. Sunshine in all cases. (After Steemami 

 Nielsen.) 



expeditions with big ships. Instead a simulated in situ method may be used. 

 In experiments using water from the different depths the samples are suspended 

 in a tube with running surface water on the deck. In order to imitate the light 

 conditions at the different depths, plates of neutral glass filters are placed above 

 the bottles containing sub -surface water. These neutral filters are chosen in 

 such a way that they match the light absorption in the sea ; since, however, the 

 light quality here changes with the depth the imitation is only approximate. 

 Beige (1958) showed that the results from such simulated in situ experiments 



