REPORT OF COMMISSIONERS OF INLAND FISHERIES. 283 



other hand, if the regeneration begins extremely late in the period, 

 the molting period may be increased in length by even as much 

 as 48 per cent., or, in other words, by nearly one-half the normal 

 period. Accordingly, since the effect of the regenerative process in 

 this experiment is fairly proportionate to the time at which it was 

 introduced, the relation may be stated as follows: the later the time 

 at which regeneration begins after the molt, the greater the length of the 

 entire molting period of the lobster. 



In estimating for comparative purposes the value of the conditions 

 involved in the experiments so far described, the following observa- 

 tion seems important. 



In the introduction to the present subject reference was made to 

 the fact that the season of the year, or, rather, the temperature of the 

 water, which is the essential condition in this instance, has a marked 

 influence upon the length of the molting period. This effect is well 

 illustrated by a comparison of the normal molting period for series 

 A with the normal period for series B, in the preceding experiments. 



The lobsters in these two series, it may be recalled, were all in the 

 same stage and in a normal condition. Accordingly it might natur- 

 ally be expected that, in both series, the molting periods would be 

 practically equal in length. But this was not the case, for while 

 the average period for series A was only 12 days, the corresponding 

 period for series B was 15.4 days; so that, instead of an approximate 

 equality in length, there is a difference of 3.4 days, or an increase of 

 over 28 per cent., in the average length of the normal molting 

 period in favor of series B. 



During the course of the experiment with series B it was noticed 

 that the general condition of the weather was cooler than it had been 

 during the previous experiment; indeed, this was so evident that it 

 suggested that this fall in temperature might account for the longer 

 molting period for series B, and accordingly a note of this observation 

 was made in the daily records. Now, since apparently the only dif- 

 ference in the conditions involved in these two series of lobsters 

 consisted in the time at which the experiments were made, it is 



