354 THE NERVOUS CYTOLOGY OF PERIPLANETA ORIENTALIS. 



and its lack of an investing membrane, will be more rapidly affected by osmotic forces. 

 It is to be remembered that all fixing agents by their chemical action precipitate 

 the protoplasmic constituents of the cell, thus rendering its structure more firm and 

 arresting the action of osmotic currents so far as altering its shape is concerned. It is 

 also to be noted that the osmotic influence of the fixing fluid is exerted before actual 

 contact between it and the cell occurs, while its chemical influence must be delayed 

 until that time, and also that the first traces of precipitation are probably flocculent 

 in character and admit of being carried inward with the mass action toward the 

 nucleus before the whole mass becomes solid. 



When the nerve-cells of the roach, then, are immersed in picroformalin, Lang's 

 solution, or chromoxalic, water is at once withdrawn from the cytoplasm on all sides 

 in a centrifugal direction and the colloid matter of the cell-body and most of the retic- 

 ulum are carried inward in a mass against the nucleus; a few radially disposed shreds 

 and bundles of the reticular fibrils remaining adherent to the connective-tissue wall 

 of the cell-space alone are left to represent the outer parts of the cell-body. It is 

 probable that precipitation now happens before the more gradual withdrawal of water 

 through the nuclear membrane and its consequent shrinkage can occur. There is also a 

 shrinkage of the supporting connective tissue through the withdrawal of water from 

 its spaces. Little violence is done to it, however, because its areolar structure per- 

 mits the water to leave it without laceration. In vom Rath's fluid for some reason 

 there is less violence from osmotic action, but the process seems to be more pro- 

 longed, for here the nuclear membrane regularly shows some shrinkage. The theory 

 that the damage done to the cell by the fluids just discussed is thus produced by 

 osmotic action is very strongly supported by the fact that Van Gehuchten's fluid, 

 not having a very high osmotic power, but possessing a strong dehydrating influence 

 from the alcohol and glacial acetic acid that it contains, causes almost identically 

 the same results as picroformalin, and that absolute alcohol, also of low osmotic 

 influence, but with still stronger dehydrating power, causes a result similar in kind 

 but greater in degree. In this instance the elements of the cell-body are more closely 

 huddled in a perinuclear ring and the water is drawn from the nucleus with such 

 rapidity as to cause the sudden rupture of its membrane, the process being too 

 sudden for the more gradual escape of the water through the membrane and 

 the consequent shrinkage of the mass of the nucleus. The gradual introduction 

 of alcohol by the diffusion-bottle caused a slow and even withdrawal of water from 

 the cell-body with a shrinkage of each fibril of the reticulum, and the diminution in 

 area of each interstice, together with distinct shrinkage of the nucleus. 



Graded formalin, on the other hand, caused an imbibition of water, distending 



