76 NINTH REPORT. 



PLANNING AN EXPERIMENT TO SHOW TO WHAT EXTENT BUM- 

 BLEBEES AID IN POLLINIZING RED CLOVER. 



W. J. Beal. 



Darwin covered 100 flower-heads of red clover b}' a net and not a single 

 seed was produced, while 100 heads growing outside yielded by careful esti- 

 mate 2.720 seeds. He concluded: "It is at least certain that bumblebees 

 are the chief fertilizers of the common red clover." 



During eight years at the Agricultural College I covered clusters of flower- 

 heads of red clover by a net, and when harvested never failed to secure some 

 seeds. Darwin's conclusion was too sweeping; it was very likely true of 

 the locality and the season in which he made the experiment. 



Suppose we wished to secure definite results as to what extent the visits 

 of bumblebees increased the yield of clover seeds. Wouldn't you select a 

 number of plants in flower near each other, cover a portion of them with 

 nets and leave a like number uncovered? There are at least two reasons 

 wh)^ such a test might not be reliable. 1. Bumblebees might be scarce or 

 absent altogether from the plants experimented upon, hence would have 

 little or nothing to do in pollination. 2. Do we know that heads of two 

 different plants would yiekl an equal number of seeds? I will illustrate this 

 second point by some experiments which I made about 25 years ago. In 

 September, five or six ]:)lants apparently nearly alike, and within a few feet 

 of each other, were selected for a test. None of them had been covered by 

 nets. Fifty heads were selected from each plant and the seeds shelled out 

 carefully with the following results: 1260, 1275, 1460, 1485, and 1820. It 

 will be seen that 50 heads from ])]ant number 5 contained about one-third 

 more seeds than 50 heads from plant number 1. At the same time in another 

 place, 50 heads selected from one plant yielded 2290 seeds, nearly twice as 

 many as were produced by plant numljer 1 in the first lot. Further, suppose 

 some one says that he thinks a net placed over the heads shading them some- 

 what is a hindrance to proilucing seeds, what can the experimenter say in 

 refutation to this criticism? 



In my later experiments made about 1882, I selected a number of thrifty 

 spreading plants, dividing the top on a north and south line, turning half 

 the top to the east and the other half to the west. Over each half independ- 

 ently was placed mosquito netting held up by stakes and spread out four or 

 five feet in diameter. Under the net covering one-half nothing further was 

 done, while under the other half several bumblebees were placed, and others 

 for several days were collected and placed inside, and not only placed inside, 

 but were seen to be busy passing from flower to flower. So far as the ex- 

 periments showed for eight years, those flowers visited by bumblebees yielded 

 about four times as many seeds as did those covered by nets. Nor are all of 

 these precautions sure to bring accurate results. We are in doubt as to what 

 extent the bumblebees placed luider the mosquito netting performed their 

 work effectually. Again clover blossoms were examined now and then and 

 small crawling insects known as thrips were seen briskly going from place to 



