14 



BULLETIN 1339, U. S. DEPARTMENT OF AGRICULTURE 



colonies was chiefly determined by factors other than internal ones. 

 The differences observed between these two colonies were almost ex- 

 clusively the actual differences of loss or gain, and the most probable 

 explanation of these differences is to be found in the presumably 

 smaller number of bees in colony 1. Actual counts of the bees of 

 the two colonies could not be taken without vitiating the experiment. 

 If any structural difference existed between the bees of the two col- 

 onies (as assumed by Merrill in his work) , this was not determined, 

 and with such high correlations as occur in the changes in weight of 

 these two colonies such an explanation seems improbable. The high 

 correlations do not indicate that any important internal difference, 

 as in the condition of the queen, or the age of the bees, existed be- 

 tween the colonies. If there had been such differences so high a 

 correlation would not have existed. It must be kept in mind that 

 .both colonies were amply supplied with storage and evaporating space. 





+ 600 



'■5 e V 8 9 /O // /2" 9 2 3 4- <5" 6 7 & P /<? // 7i? '/ 2 3 4 3 s 



Fig. 5.— Graphs of the average variations, hour by hour, in the weight of Colonies 1, 2, and AB far the 

 May honey flow. Note the minimum of the midday decline at 2 o'clock for the three colonies 



If one of the colonies had been crowded for space, and this had inter- 

 fered with the work of gathering, undoubtedly the coefficient of 

 correlation would have been smaller. 



An examination of these data indicates that the field forces of the 

 two colonies were working at approximately maximum efficiency. 

 The two colonies gathered a total crop for the year 1923 which com- 

 pares favorably with that of any other colonies kept in the general 

 locality. An examination of the striking similarity of the changes 

 in weight of the two colonies at the time when both regained their 

 morning weight and at the time of the midday decline of increase in 

 weight strongly suggests that both colonies were gathering all the 

 nectar which was available for them, in proportion to the number of 

 l>ees available in each for field activity. This similarity is brought 

 out much more clearly in the averages for the entire May honey flow 

 (fig. 5) than in those for individual days (Table 1) , since toward the 



