EXPERIMENTS ON FIXATION 319 



(A smaller volume should be used when osmium tetroxide is the 

 fixative.) Drop a prepared pipette into the specimen-tube. It will 

 float, bulb uppermost. Note the time. The fixative will begin to 

 penetrate into the gel from the bottom of the tube. 



Make observations at intervals on the distance penetrated by the 

 fixative. It is best (though rather inconvenient) to make them at 

 4, 9, 16, and 25 hours after the fixative has begun to act. 



The method of measuring the rate of penetration of the fixative 

 depends on whether the latter is coagulant or not. If it is, simply 

 note where the coagulation has reached, measure the distance 

 from the diamond-mark to the limit of coagulation, and subtract 

 this distance from the measurement made at the beginning of the 

 experiment. The limit of coagulation by most fixatives can easily 

 be seen, on account of the resulting opacity. (It is rather difficult 

 to see the limit when the fixative is hydrochloric acid or mercuric 

 chloride, unless one is careful to arrange suitable lighting and 

 background.) 



A more complicated procedure is necessary with non-coagulant 

 fixatives, such as formaldehyde, because one cannot see the limit 

 between the fixed and unfixed parts of the gel. To determine the 

 position of the limit, remove the rubber bulb from the tube, put 

 it on the other end, and then float the pipette in water at 37° C. 

 In a few minutes the unfixed gel will have run out of the tube, and 

 it will then be obvious how far the fixative has penetrated. It is 

 necessary to use a separate pipette for each observation with 

 each fixative. 



The method described in the preceding paragraph is applicable 

 also to osmium tetroxide. The fixative produces a black gel at the 

 bottom of the tube, rather sharply marked off from a brownish 

 yellow region above, which fades into the unaffected gel at the top. 

 It is easy to show that the part of the gel that has been blackened 

 is the only part that has been rendered insoluble in water at 37° C, 

 and the limit of the blackened region may therefore be used as an 

 index of the distance penetrated by osmium tetroxide at fixative 

 strength. 



It is to be noticed that what is measured is the thickness of the 

 layer of the original gel that has been penetrated by the fixative at its 

 effective concentration. This is by no means necessarily the same as 

 the thickness of the fixed gel, for many fixatives shrink or swell it. 



The results of the experiment are show^n in fig. 3 (p. 39). 



Acetic acid and potassium dichromate do not coagulate the 



