CHANGES ]N ENVIRONMENT OF DAPHNIA MACiNA. 



219 



remaining two were alive but apparently sickly at the middle 

 of December. 



That this result was due to the unavoidable shaking in 

 railway travelling would seem to be unlikely. It is con- 

 ceivable that the sudden stimulus produced by transfer from 

 one water to another was prejudicial, and that the second 

 application of a like stimulus was still more damaging to the 

 race. 



The following table shows that the two kinds of water 

 differ widely in the amount of their mineral constituents. 



Canterbury Water.' 



Grains per gall. 



Carbonate of lime 

 Suli)liate of lime 

 Nitrate of lime . 

 Mai^nesia . 

 Alkaline ciilorides 

 Silica .... 



Total 



1-39 

 0-07 

 2-04 

 0-27 

 3-41 

 0-40 



7-58 



New River Water.'^ 



Grains per i 

 Carbonate of lime . . 12-70 

 Sulpliate of lime . . . 1"G0 

 Nitrate of lime . . . l-QO 

 Nitrate of mas;nesia . . 1'28 

 Chloride of sodium . . 2'02 



Silica 0-26 



Alumina, etc 14 



Totj 



1900 



Thus in the New River water there is two and a hall: times 

 more mineral matter than in Canterbury water. 



The kind of reaction which was observed is very similar to 

 that which M. W. M. Hafkine ^ found among two races of 

 Infusoria. A race living in a clear pond near Paris was 

 killed by an artificial infusion of hay and leaves in which 

 the same species of Infusoria was flourishing. 



y. Reproductive Cycles. 

 Another possible explanation may be sought in the follow- 



1 This analysis was kindly sent to me by Mr. S. Harvey, the City 

 analyst. 



' The analysis of the New River water is taken from Dr. J. A. Wanklyn's 

 ' Water Analysis,' 1896. 



' "Recherches sur I'Adaptation au Milieu chez Infusores et les Bacteries," 

 par W. M. Hafkine, 'Ann. lust. Past.,' 1890. 



