The Ilominfj of the Ilymenoptera 43 



black and white transverse stripes; and some speckled 

 with red and green. These colors were not optically 

 pure, but were such as could be procured in the market. 



The place selected for the experiments was a large 

 field of white, sweet clover, which was visited by count- 

 less numbers of bees. The experiments occupied a part 

 of two summers. The bees were trained to visit an arti- 

 fact of a certain color by placing honey on or in an arti- 

 fact of that color and leaving the others empty. The 

 first season they were trained to visit a plain artifact; 

 the second, to visit a striped one. Much time and patience 

 was required to induce a single bee to visit one of these 

 artifacts. Soon after one of the bees had discovered the 

 richness of these artifacts in honey, a large number of 

 bees began to make regular visits to them. After the 

 bees had been foraging upon these artifacts for several 

 days and had practically neglected the melilot for them, 

 new artifacts were constructed and the following exam- 

 ination was held : 



The tests covered a number of days and were arranged 

 in series. In one series no honey was placed in any of 

 the artifacts; in another, honey was placed in some of 

 the artifacts, but none in those of the color and pattern 

 from which the bees had been trained to expect honey ; in 

 a third series honey was placed in some of the artifacts 

 of each color pattern used. The number of varieties of 

 color patterns used at one time varied from two to seven 

 and the total nmuber of artifacts used at one time vai'ied 

 from two to twenty. Almost invariably the bees flew 

 into the artifacts of the color and pattern from which 

 they had been trained to collect honey ; but paid no atten- 

 tion to artifacts of any other color and pattern. The 

 manner in which the bees passed by most artifacts, even 

 when they contained honey, and the eagerness with which 



