POLLEN LOADS. 



119 



series as well as between the two calendars made at the same time, as shown 

 in table 86. 



The greatest difference is shown by the two calendars of series 3, in which 

 the marked constancy of Clisodon and Bombus appositus and their vastly 

 greater abundance in the one place furnishes the explanation. Corre- 

 sponding differences may also be noted with respect to the response to the 

 several competitors. Normal Geranium flowers and composite heads of 

 Chamaenerium were twice visited equally, though as a rule these artificial 

 heads were neglected. A marked exception occurred in the second series, 

 where 14 flowers of a pure Gilia head were visited by Argynnis. As between 

 normal Chamaenerium and normal Geranium, the former was thrice as 

 attractive all told, though the relative positions were reversed in the cal- 

 endars of the second series. In this competition B. bifarius and B. hunti 

 were partial to Chamaenerium and B. occidentalis to Geranium, while B. 

 juxtus preferred Geranium in one series and Chamaenerium in the other. 



Table 86. — Summary of normal and competing flowers by series and calendars. 



COMPOSITION AND WEIGHT OF POLLEN LOADS. 



Value and methods. — While occasional observations of the purity of 

 pollen loads have been recorded (Bennett, 1874; Perez, 1903), no previous 

 microscopical analysis of loads seems to have been made. The latter were 

 undertaken in the first place as a check upon the habits observed in the 

 field, and were extended to include the weight of the bee and its pollen load 

 as a measure of its efficiency in collecting and in flight. Unfortunately, it 

 was not until these preliminary studies were practically finished that it was 

 realized that the time of flight should have been recorded, since the degree 

 of constancy varies much between morning or evening on the one hand 

 and mid-day on the other. There is also an evident difference in indi- 

 vidual efficiency, which can be determined only by the use of marked bees. 



As a rule the bees were weighed immediately after being caught, the 

 load completely removed by means of a needle, and the body weighed 

 again. In a few cases the bees were dry, and corresponding allowance must 

 be made with respect to the percentage weight of the load. In the tables 

 the name of the species on which the insect was captured is given first in 

 each case; when the bee had collected none of this pollen, it is indicated. 

 For a number of specimens previously collected (designated by a letter), 

 the pollen was examined, but the weight of the load was not determined. 



