300 B. O. Peirce—Sensitiveness of the Eye to Color. 
In no case was the seh able to detect any difference 
between the colors of the two edges of a strip; each strip 
_ seemed to be of one color saa 
It was necessary to regulate very carefully the Argand 
burner used as a source of light, for the least difference in the 
intensities of the two spectra made the strip of the brighter 
seem to the phere yellower than the corresponding strip i) 
the other spectrum. This error, due to the “color of bright- 
_ ness,""* was most apparent in observations gt at the extremi- 
ties of the spectrum, i. e. in the red and the v 
A displacement of the movable collimator it ‘amounting to 
: 1 scale desis corresponded to 4 minutes of arc, or to a differ- 
oe ence in Reniscstlee of -0000013™™ 
eS rver sat in a darkened room and directed the tele- | 
_ scope fe that part of the spectrum in which the observations 
were to be made. After an assistant had placed the movable 
- collimator slit even with the other, and the observer had seen 
to it that under these circumstances the strips of light in the 
_ eye-piece seemed just alike—as they should—the observer took 
his eye away from the instrument, and the. assistant displaced 
the movable collimator slit and asked the observer to look into 
: the telescope and to tell him, from the colors of the strips, in 
wh n 
— with the fixed slit. If the displacement was very large, _ : 
‘slit i se order to match the colors of the strips; if in successive 
trials the displacements were made smaller and Sadatler a place 
was reached where the colors of the strips did not seem well © 
rver was able to tell immediately how to move the : : 
matched to the observer, but where he could no longer say _ 
2h ase as in which direction the movable slit had been a 
first somewhat irrega ile y 4 
os ails were invariably obtain 
same person, even though the experiments were sepa 
te oy days or weeks, and that certain features were eens on 
to all the series obtained with the cle evita observers : 
The la largest acement needed (except at the ends. of the 
trum) b, y any of the observers corresponded to difference 
mgth of the midd] e of the two strips of about *000000" 
