4 BULLETIIN- 804, IT. s. DEPARTMEITT OF AGEICULTUEE 



nor healthy unsealed larvae to receive the infection, and the disease 

 will thereby come to an end. 



Dr. Miller has been using a 10-day period of queenlessness in his 

 treatment of European foulbrood since his accidental discovery that 

 10 days were sufficient, but in a later article (3) in enlarging upon 

 his nurse-bee theory he assumes that the larva is fed during a period 

 of 5 days but is not effective as a carrier of infection during the whole 

 time as probably no larvae are torn open until they are 2 or 3 days old, 

 thus making it possible to shorten the queenless period even more. He 

 admits that not all the dead, partially dried larvae will be cleaned out, 

 but believes that it is only the fresh yellow ones which are infectious. 

 He also states that nurse bees are not inclined to travel far on the 

 combs, a fact which may explain why the disease may be found con- 

 fined to one comb for several days before spreading farther. Dr. 

 Miller seems to have overlooked several important factors which 

 will be discussed later. 



Quite an extensive piece of investigation was carried on during the 

 summers of 1915 and 1916* by the author at the Massachusetts Agri- 

 cultural Experiment Station upon the effect of requeening diseased 

 colonies with various strains of Italian bees. At that time the im- 

 portance of strong colonies with the requeening had not been em- 

 phasized so strongly and less attention was paid to that factor. The 

 records show, however, that in a total of 50 colonies observed, cov- 

 ering two seasons, of 10 strong colonies only 2 showed recurrence, 

 while 1 was doubtful ; of 20 medium-strength colonies, 10 showed re- 

 currence with 2 doubtful ; of 14 weak colonies, 8 showed recurrence. 

 In all these cases the new queen was not introduced until the colony 

 was nearly or entirely clean. In the case of several of the weaker 

 colonies it was necessary to strengthen them before requeening was 

 possible, in order to save the colony. One or two of these, which were 

 united and requeened with Italian stock, were the best colonies the 

 next spring. 



Adding some strength to at least part of Miller's theory is a state- 

 ment in a letter by G. C. Matthews, formerly of this bureau, who 

 wrote in February, 1918, concerning his observations in California 

 in 1911. He found that where the hives stood in rows of pairs the 

 disease continued to spread down each row to corresponding members 

 of each pair. This ceased when he rearranged his apiary so that 

 the rows of hives were at least 10 feet apart, and alternate pairs 

 of hives were turned at right angles. No pair was allowed to remain 

 close to another facing the same way. This prevented the drifting 

 of nurse bees, which he believes to be the method of spreading the 

 disease. Furthermore, he found by introducing one Italian queen 

 into the middle colony of an isolated row of hybrid bees that there 

 was considerable drifting of nurse bees. Seven days after the brood 



