REPORT OF COMMISSIONERS OF INLAND FISHERIES. 285 



A comparison of these temperatures shows that during the first 

 part of July the water was much warmer than for the last part of 

 July and the first part of August, so that there was a difference of as 

 much as 5° in the average temperature for these two periods in favor 

 of series A. 



This comparatively large difference in temperature undoubtedly 

 accounts, to a very large degree, for the corresponding difference in 

 the normal molting periods for the two series of lobsters. This fact 

 very clearly demonstrates the importance of securing similar condi- 

 tions in season and temperature for experiments of this nature in 

 order to obtain the most reliable data for comparative purposes. 



Again, it may also be observed in the above temperature readings 

 that, during the latter part of July, there was a gradual decline in 

 the temperature of the water. Now, since in the experiments with 

 series C4 to C^g the last molts occurred at late as August 13, it 

 might be suggested that, possibly, the delay in molting in these 

 specimens may have been partly due to a decline in temperature 

 as well as to the process of regeneration. That this could not have 

 been the case, however, may be readily seen from the following 

 temperature readings of the water up to August 13 : 



August 1 68° 



August 2 68 . 5° 



August 8 72° 



August 9 74° 



August 12 7! 



TOO 



From these readings it will be seen that for some climatic reason, 

 instead of a decline, there was a marked rise in the temperature 

 of the water from August 1 to August 13. 



This marked increase in temperature during the latter part of the 

 molting period for series C4 to C^g would evidently tend to hasten 

 rather than retard the rate of molting. Accordingly, the tempera- 

 ture conditions in the latter part of this experiment, instead of 



