BUCKINGHAM. — DIVISION OF L.\BOK AMONG ANTS. 453 



every 24 hours, during at least one series when the food was in A or in B. 

 A comparison of the results of these two tables shows that there is one 

 colony (51) from which during the 24-hour periods more ants entered B 

 when the food was in A than when it was in /i ; but which in the 48- 

 hour periods gave the opposite result. All those colonies which in the 

 24-hour periods showed more ants entering /y when the food was in B 

 than when it was in A, showed the same results in the 48-hour periods. 

 In the 48-hour periods, as in the 24-hour periods, when there was more 

 than one series with food in B, the per cents of ants in B were fairly 

 similar in the two cases (84.3 and 28 ; 6 and 4.6), nor are the results very 

 different in a given colony (17 and 20) for the 24- and 48-hour periods. 

 On the whole, when we compare the results in the 24-hour periods and 

 the 48 -hour periods, it will be seen that there were somewhat more ants 

 entering B during the 48-hour periods than during the 24-hour periods. 

 In the 48-hour periods there were rather more ants entering B when 

 the food was in A than when it was in B, as opposed to the reverse 

 condition shown for the 24-hour periods. I believe this to be due to 

 the fact that there were several colonies (4, 17, 20, and 24) which, 

 throughout one or more series, were examined daily (consequently no 

 48-hour periods) and these were colonies which, during the 24-hour 

 periods showed more ants entering B when the food was in B. 



To ascertain the constancy of individuals in regard to foraging, I noted 

 the number of times which each ant had the opportunity (one opportun- 

 ity in 24 hours) to go into B, and also the number of times which it 

 availed itself of this opportunity. From the ratio of the two numbers, 

 expressed in per cents, was subtracted a similar ratio ascertained when 

 the food was in A, since we must assume that the ants entering B 

 when the food was in A did so for some other purpose than that of 

 obtaining food, and that an equal number would presumably have 

 passed from ^ to ^ when the food was in B for similar reasons, i. e. 

 not for food. These results are given in Table III, where the second 

 column shows the number of ants engaged in foraging for each of the 

 several per cents from — 78 to 85. For the sake of convenience the re- 

 sults of the corresponding operation for the activities of tending the 

 young (third column) and building (fourth column) are incorporated 

 in the same table. The results on foraging would have been more 

 convincing had more ants in all colonies entered B when the food was 

 in B than when it was in A. But as has been stated, it was frequently 

 found that an individual ant entered B more often when the food was 

 in A than when it was in B. 



In compiling Table III, I have disregarded in respect to every activ- 

 ity all individuals which died before the end of the experiment. There 



