18'4 FIFTY .YEARS AMONG THE BEES 



it may happen in some seasons that preparation for swarming 

 begins during the hist of fruit bloom. Of late years dandelion 

 has become so imijortant that Ihei'e is a possibility it may start 

 swarming. Whether it be in apple or clover bloom, we begin 

 to examine some of the strongest colonies to see if any prepar- 

 ations for swai-ming aie made. If we find none in the strong- 

 est colonies it is hardly woi'th while to look through the rest. 

 When, however, ^\e find one or more queen-cells witli an egg 



/'((/, (id — Top and Bottom Starters in Section, 



in each, tlien it is time to begin a systeuuitic canvass of all 

 colonies, and to keep it up in all so long as we continue to find 

 queen-cells in any, except in a case where a colony has already 

 been treated or has treated itself in sucli a way that it need not 

 be expected to swarui. 



COI.ONIES THAT DO NOT NEED WATCHINa. 



Tn struggling with the swai-ming problem, there are a few 

 things that umy be relied upon with some degree of certainty. 

 A swann that has been hi\eil in nn enqity hi\e this season will 

 not senil forth n swarm tliis year, with rare exceptions. Equally 

 safe from swarming is a colony whose queen has been removed 



