510 S. O. MAST 



tests were made; the opaque shades were frequently raised and 

 lowered and the artificial light in the room was also repeatedly 

 turned on and off. From three until four o'clock, however, 

 the shades were continuously up and during this time the light 

 was also stronger than it had been earlier in the day. After 

 being thus exposed for an hour colonies taken from the jars were 

 again tested and now, after having given very indefinite re- 

 sponses all day, they were, to my great surprise, very strongly 

 positive and oriented remarkably precisely, and they continued 

 to respond thus during a long series of tests, the results of which 

 are among the most accurate and consistent obtained. They 

 are given in columns 2-5, table 10. 



In these tests fresh colonies were taken after every second 

 reading. One pipette-full of tap water and onerhialf pipette 

 full of w^ater from the jars containing numerous colonies were 

 put into the observation aquarium, then a reading was made in 

 a given color and the result recorded in column 2. The solu- 

 tion in the observation aquarium was now thoroughly stirred 

 and another reading made in the same color and the result re- 

 corded in column 3, after which the wave-length was changed by 

 10 ixn, fresh colonies put into the aquarium and the process re- 

 peated. All of the readings in this series of tests were made 

 in this way, and thus variation in the degree of deflection was 

 largely eliminated. This is shown by the remarkably close 

 agreement in the results obtained at different times in the same 

 wave-lengths and recorded in table 10. 



In this table are also recorded the results obtained in two of 

 the best series of observations made with colonies in the negative 

 state. By examining this table, table 15, and figure 4, in which 

 some of the results are plotted, it becomes evident at once that 

 the distribution of stimulating efficiency in the spectrum is 

 strikingly different for Pandorina and Euglena, Chlamydomonas 

 or any of the other forms previously discussed. The region of 

 maximum efficiency is considerably nearer the red and the 

 effective part of the spectrum extends from the maximum in 

 either direction much farther. The curve representing the re- 

 sults obtained with Pandorina is consequently considerably 



