ECOLOGY. 341 



that it will afford a check on the photographic method, and a new colori- 

 metric photometer that makes use of chlorophyll is being developed. A 

 paper is now in preparation dealing with the series of photographic pho- 

 tometers and the spectro-photometer and their uses. 



Methods and Principles in Experimental Pollination, by F. E. Clements 



and Frances Long. 



In a final extension and checking of the results of the past 5 years, further 

 experiments have been made as to the response of insects to color, odor, muti- 

 lation, and artificial flowers and to determine the relative attraction of different 

 species when competing with each other. Life-histories have again been 

 traced in detail, and a method of observation and record has been devised for 

 the florets of composites and grasses. The observations on the normal 

 behavior of the principal pollinators have been extended, and the number of 

 pollen-loads weighed and analyzed has been greatly augmented. A particular 

 endeavor has been made to follow the behavior of Bombus, especially during 

 the interval between the maximum flowering of two preferred species. In 

 connection with the relative efficiency of flowers as to pollen deposit, many 

 permanent mounts of pollen-grains and stigmatic surfaces have been made. 



The summer's work has confirmed the general conclusion, already reached, 

 that habit is the most important factor in the behavior of pollinators, though 

 the influence of habit varies with the group, species, age, season, time of day, 

 and other considerations. Contrary to Plateau's views, color appears to be 

 more controlling than odor, except in flight from a distance. In the effort to 

 give greater currency to the experimental study of pollination, an exhaustive 

 summary has been made of all previous researches that have made use of 

 experiments, and detailed attention has been given to the complete work of 

 Plateau and his critics. While it is planned to extend the investigation into 

 the alpine climax and to the Pacific Coast next summer, the results of the 

 present series of experiments are now being brought together for publication. 



Factors Involved in the Opening and Closing of Flowers, by G. W. Goldsmith 



and I. M. Johnston. 



In connection with pollination experiments, it has proved desirable to 

 scrutinize the results obtained in earlier investigations and to initiate a new 

 series of studies to deal more accurately with the factors and conditions 

 concerned in the field. In the case of Mentzelia multiflora, the flowers open 

 regularly between 4 and 5 p. m., the younger flowers closing about 8 o'clock 

 and the older ones remaining open until the following morning. When 

 plants were placed in bell-jars heated by means of a steam-coil and the 

 temperature kept at 25° C, the time and speed of closing were not perceptibly 

 changed, even though the temperatures were maintained for several hours 

 before the time of opening. However, it was possible to prevent closing by 

 such means, while the removal of the bell-jars usually resulted in closure within 

 a half hour. Thus, a fall in temperature seems to be responsible for normal 

 closing, but opening appears to be independent of the usual changes in tem- 

 perature and is apparently related to turgor. If stems are cut and the ends 

 placed in water, there is usually no movement for the next 24 hours, regardless 

 of changes in temperature or light. Closed flowers fail to open, and open ones 

 do not close until the following day, unless the water is under approximately 



