572 Wisconsin Academy of Sciences, Arts, and Letters. 
There are considerable differences between the results reached 
by these authorities. The observations of Hiifner and 
Albrecht, Ewan, and Aschkinass were all made between 1891 
and 1895 and had the direct purpose of ascertaining the ab¬ 
sorption of light by water. The investigation of Aschkinass 
is the most complete and extensive and his results are probably 
the most accurate. The work of von Aufsess on distilled water 
was done in preparation for studies on lake water. It will 
be noticed that the results of von Aufsess are regularly lower 
toward the blue and higher toward the red end of the spectrum 
than those of Aschkinass or Ewan. My own determinations 
are in quite as close agreement with those of Aschkinass as 
could be expected from methods so diverse. 
Hiifner and Albrecht were the first to obtain results for dis¬ 
tilled water. They compared the intensity of the incident 
with the transmitted light which had passed through a glass 
tube two meters in length, filled with water. The width of 
spectrum used was large, covering 100 Angstrom units or more. 
Ewan used a method by which two lenses were filled with 
light from the same slit, illuminated by an incandescent Auer 
lamp. One beam of light was passed through a glass tube 
one meter in length, containing the water, and then compared 
with the other beam as a standard by means of a Lummer and 
Brodhun prism. The variation of intensity of the standard 
beam was produced by means of a rotating sector. The water 
first used was distilled twice so it would be as free as possible 
from dust. He found afterwards that water filtered through 
a Chamberland Pasteur candle was even more devoid of dust 
than that which had been freshly distilled. Measurements on 
such filtered water however showed no noticeable difference 
in absorption. 
Aschkinass determined the absorption-spectra of water 
chiefly for the infra-red rays. The instrument used was a 
bolometer of ordinary construction and the layers of water 
varied from .001 cm. to 100 cm. in thickness. In the infra-red a 
fluorite prism was used, while in the visible spectrum a flint glass 
prism and, due to the small amount of energy, a larger 
spectroscope and a more delicate bolometer were required. 
Nothing is said in regard to the purity of the water. 
Observations were made by 0. von Aufsess with the aid 
of a Nernst lamp, a Konig-Martens spectrophotometer, and a 
