Cross — Eupatorium ageratoides and coelestinum. 261 



and completely covered in by the involucre. The remain- 

 ing four plants were marked with bits of muslin and left 

 exposed. Six days later, twelve additional plants from each 

 species were selected. Two of these were covered by thin 

 cotton cloth, two with Swiss muslin, two had one-half of 

 each inflorescence protected by thin cotton cloth, and the 

 other half exposed, two had one-half of each inflorescence 

 protected by Swiss muslin, two had single branches of 

 the panicle protected, and the remaining plants were left 



exposed. 



In each lot of plants selected, those remaining uncovered 

 so developed their flowers that the style arms divaricated 

 about three days earlier than those covered with muslin. The 

 mnslin-covered plants matured their flowers two days earlier 

 than did those covered with cotton cloth, this difference 

 being probably due to the greater amount of light passing 

 through the open meshes of the Swiss muslin, than through 

 the more closely woven cotton cloth. The difference in 

 the time of maturing was also very noticeable in the plants 

 which had a part of the inflorescence protected and a part 

 exposed, both having similar conditions, except in the 

 amount of light. Uncovered parts were faded, when covered 

 parts were just maturing. Another interesting difference 

 between protected and exposed flowers, whether on the 

 same or separate plants, was that in every instance the 

 protected florets had their style arms developed in a very 

 contorted manner. Those of one floret often touched 

 a neighboring style or dipped into the adjacent corolla. 

 Hildebrand states that the style arms of chicory roll up 

 like a feather, the pollen being thus brought in contact 

 with the stigma. This he considers to occur most fre- 

 quently in the absence of insects. It was difficult to col- 

 lect a small amount of pollen from exposed plants, but 111 

 protected ones the pollen was so abundant that it covered 

 the inflorescence and the inside of the muslin cover. 

 Indeed, it is difficult to form an adequate conception of the 

 amount of pollen produced unless the inflorescence be pro- 



