158 



PRINCIPLES AND CONCLUSIONS. 



Table 91. 



Species. 



No. of 

 individuals. 



No. of 

 errors. 



Proportion per 

 individual. 



Apis mellifica 



Bom bus hortorum 



muscorum 



lapidarius 



Megachile centuncularius 



ericetorum 



Odynerus quadratus 



Total 



107 



186 



1.8 

 1.5 

 2.2 

 1.0 

 1.0 

 1.0 

 1.0 



Removal of the antennae of bumble-bees. — In the introduction to 

 this paper (1902 2 :414), Plateau stated that up to that time he had attributed 

 a secondary role to sight and a preponderant one to smell, and likewise 

 called attention to the results of many investigators, including Forel and 

 himself, who had shown that with the insects the sense of smell resides 

 chiefly if not exclusively in the antennae. Forel (1901:53) cut the antennas 

 of 6 individuals of Bombus at the base and set the bees free again. At the 

 end of 5 minutes a male returned to visit 8 to 10 flowers of Convolvulus 

 in succession, each time flying directly to the flower without a second's 

 hesitation. He was caught to confirm the complete absence of the antennae 

 and then released, when he made a single turn in the air and came back 

 at once to the flowers to visit them as before. At the same time, several 

 of the bumble-bees deprived of their antennae returned to the bindweed 

 flowers, flying from one to the other with even more precision if possible 

 than those with the antennae intact. Two days later several males and 

 small females with the antennae cut were found flying from flower to flower 

 with an astonishing rapidity and precision. 



Plateau repeated the experiments of Forel on two occasions. In 1899 

 the antennae were cut from 10 individuals of Bombus terrestris and 1 ot B. 

 muscorum, on four nearly successive days. These at once flew away, but 

 one returned the first day to the same flowers, and on the following days, 

 and in 3 cases shortly after the operation, 6 bees without antennae were 

 captured, including the B. muscorum. Although more than half of the 

 multilated bees had found their way back to the flowers, the experiment 

 was repeated in 1902 to remove certain doubts, especially with respect to 

 the thoroughness with which the antennae were removed. The latter were 

 consequently cut at the level of the head and the bees released. The results 

 were as follows: 4 neuters of B. terrestis never returned; of 11 B. hortorum, 

 10 females and 1 neuter, only 1 female returned; 1 male and 1 female came 

 back out of 7 individuals of B. lapidarius; finally, of 8 B. hypnorum, 5 re- 

 turned shortly after the operation. Thus, 8 bumble-bees out of 30 returned 

 accurately to the preferred species of flower after the antennae had been 

 completely removed and guidance by smell was impossible. 



