262 
PATHOLOGY: W. N. BERG 
Into each of 2 erlenmeyer flasks of 200 cc. capacity, pipet 50 cc. por- 
tions of the serum. To each flask there were added 25 cc. of water 
and 32 cc. of saturated ammonium sulfate solution; resulting in 30% 
saturation. At this concentration euglobulin was precipitated. One 
flask, not to be heated, was stoppered and set aside. The other flask, 
to be heated in the water bath, was stoppered with a rubber stopper 
carrying a standard thermometer. The bath had been so adjusted 
that during the experimental heating of the serum mixtures, the bath 
temperature did not rise beyond 61? 2 nor fall below 60 ?5. The tem- 
peratures inside the flasks (it was convenient to work with 2 sera at a 
time) rose from that of the room, about 27°, to 56° in the first five min- 
utes of the heating, then to 59° in the next five minutes. After ten 
minutes heating the temperatures inside the flasks were exactly 60° 
or below it by a small fraction of a degree. The heating was then 
continued for exactly thirty minutes during which time the tempera- 
tures inside the flasks did not exceed 60° At the end of the heating 
period the flasks were transferred to a pan containing cold water. This 
brought the temperature down to that of the room, in about five min- 
utes. The next step is the separation and estimation of the precipitat- 
ed euglobulin. 
Method of analysis. — All four serum mixtures, 2 heated and 2 not 
heated, were transferred to 110 cc. centrifuge tubes and centrifuged for 
twenty-five minutes at about 2500 revolutions per minute. The sedi- 
mentation was perfect; the sticky euglobulin was firmly packed at the 
bottoms of the tubes. The supernatant liquids containing the pseudo- 
globulin, albumin and the antitoxin, were poured into 100 cc. volumetric 
flasks. The tubes could be inverted without loss of precipitate. The 
euglobulin was dissolved in water, coagulated by heat, filtered, dried 
and weighed by methods already described. ^ In the supernatant 
liquid estimations were made of (1) pseudoglobulin ; (2) albumin and 
(3), these two together. 
Results. — The figures in table 1 show unmistakably that more euglob- 
ulin precipitate was obtained from the heated than from the unheated 
serum. The figures for euglobulin in table 2, item (b), are the same as 
those in table 1, divided by 5. The main object of obtaining the other 
data was to ascertain the source of the excess of euglobulin in the heat- 
ed precipitates and to ascertain how accurate or inaccurate the method 
of analysis was. Theoretically, in table 2, (a) should equal the sum of 
(b) plus (c), and (c) should equal the sum of (d) plus (e). Although this 
equality is absent, the differences appear to be consistent and indicate 
a uniformity of error due to the absence of corrections for the compara- 
