ACONITUM COLUMBIANUM. 



19 



of Bombus to the two colors is shown by table 3, which records the number 

 and order of visits, each time given representing a different bee. 



During the 40 minutes, 9 bees visited 75 flowers, of which 26 were white 

 and 49 purple. On another day, 63 flowers each of purple and white were 

 observed for half an hour. During this period Bombus juxtus visited 64 

 purple and 68 white flowers, and B. edwardsi 1 purple and 3 white ones. The 

 total number of visits was 65 to the purple and 71 to the white. 



Artificial flowers. — To obtain further evidence as to color preference, 

 other colors were tried. This was done in two ways: (1) by replacing 

 the corollas and calyxes with crepe-paper floral envelopes, leaving stamens, 

 pistils, and nectaries intact, or surrounding the natural corollas by crepe- 

 paper disks, and (2) by painting the perianth with water-colors. Artificial 

 corollas and calyxes of crepe-paper were made to resemble those of the 

 natural flowers as nearly as possible, the effect being essentially lifelike. 

 Yellow, green, white, red, pale blue, and purple were the colors used. The 

 artificial corollas were not attached until after the bees had visited all 

 the flowers under observation, in order to make sure that nectar was pres- 

 ent and that nothing about the flowers was objectionable. In placing the 

 artificial flowers the following combinations were used: 



(1) A colored crepe flower was paired with a normal one at various points on the stem. 



(2) A raceme was left with as many normal as crepe flowers of the various colors, and 



these were scattered in no definite order. 



(3) Flowers of a single color of crepe-paper, but as numerous as the normal flowers, 



were scattered at various points on the stem. 



(4) One half the raceme was left normal and artificial corollas of one color of crepe- 



paper were used on the other half. 



These experiments were tried on several days during the season for four 

 different years, with essentially the same results. Bombus would fly 

 within 2 cm. of the crepe flowers, but would then pass without landing. 

 The only two exceptions to the above were the flowers with the pale-blue 

 and white perianths. Bombus stopped at the white ones and worked as 

 if noting no difference, while pale-blue flowers were visited at 7 different 

 times by as many individuals of juxtus. In two cases after visiting the 

 flower and taking nectar, this species returned for another visit. 



