198 rROCEEDIXGS OF THE ROYAL SOCIETY OF QUEENSLAND. 



continu(> to live in A\ater where the COg content averaged as 

 high avS G ec. per litre throughout the year, but that it was still 

 to be demonstrated whether there were any species of truly 

 freshwater fish which could reproduce Buceessfully in water 

 that was decidedly alkaline to phenolphtlialein throughout the 

 year. Shelford (1918c, pp. 45-6) pointed out that since COg 

 results from the decomposition of organic matter, in the process 

 of which oxygen is consumed, so the presence of any large 

 quantity of CO., nearly always indicates a lack of oxygen. He 

 thinks it probable that the CO 2 content should not average 

 more than 3 cc. per litre over breeding grounds, and more than 

 6 cc. per litre during the summer, as such quantities are not 

 usually accompanied by lack of oxygen. He suggested that the 

 amount of CO2 might be taken as an index of the suitability of 

 the water for fish-life." Wells (1915) found that the COg 

 optimum for the various species of freshwater fish experimented 

 upon under summer conditions varied from the acid side of 

 neutrality to 6 cc. per litre. Marine fish behaved differently as 

 they preferred slight alkalinity to acidity (Shelford and Powers, 

 1915; Shelford 1918c, p. 40 ; 1919). The time taken to kill 

 freshwater fish, using higher concentrations of acid, was found 

 by Wells to be proportional to the hydrogen ion concentration.^ 

 This author gave considerable attention to the reactions of fish 

 to the ions of H and OH.^ 



* See also Birge and Juday (1914, pp. 583-7) regarding the 

 distribution of COg in lakes. Also Shelford, 1918c, pp. 40-1 ; 1914. 



^ " The theory of solution explains acidity in water by the occurrence 

 of hydrogen ions, formed from dissolved electrolytes, in excess of hydroxyl 

 ions ; and alkalinity by a similar excess of hydroxyl over hydrogen ions. 

 Neutrality is, then, the condition when, as in pure water, the two concen- 

 trations are equal." (L. J. Henderson, The fitness of the environment. 

 MacMillan, New York, 1913, p. 142.) 



^ By titration, using phenolphthalein an d methyl orange as indicators, 

 Wells (1915) determined the amount of COg present in a fixed condition 

 (as carbonates), " half bovind " (I.e. bicarbonates), and .free (as COg), 

 since methyl orange remains unaffected by carbonic acid so that the 

 bases present as carbonates or bicarbonates can be titrated with an acid ; 

 while carbonates are alkaline to phenolphthalein, bicarbonates neutral, 

 and free CO, acid. Methyl orange is very sensitive to OH ion whereas 

 the latter indicator reacts to the H ion instead and consequently gives 

 an acid reaction with COj- In the presence of COg, methyl orange will 

 give an alkaline reaction though the water may still be acid owing to the 

 presence of a higher concentration of H than OH ion. — See also Shelford, 

 1919b. 



