16 Alfir(/ J. Eivurt: 



The same experiinont as above was repeated, but the prepared 

 spots were surrounded by a square centimetre paraffin cell 2 — 3 mm. 

 deep through which solution trickled (see PI. 3). 1 litre was used 

 and trickled once daily through the cells and over three prepared 

 spots for seven days. 



1 per - Slight superficiitl - After 7 days in air super- 



100,(XX),000 browning ficiai browning extend- 



ing up to 2 mm. deep. 



1 per - No distinct signs - Faint to distinct superficial 



],0C0,000,000 of poisoning browning. 



1 per - No signs of p!>i- - No signs of poisoning. 



10,000,000,000 soning 



Even superficial browning under the microscope can be seen to 

 affect several successive layers of cells, one below the other, and even 

 in small pits the number of cells affected soon runs up to the 

 thousands. Since it must take a definite number of molecules even 

 of mercuric chloride to completely poison each cell, there must be a 

 limit to the dilution at which a perceptible poisonous action can be 

 exei-cised. 



A litre of a 1 per 10,000,000,000 solution ccnitains ^^"(m °^ '^ 



milligram of mercuric chloiide, representing some —^ individual 

 molecules, and assuming that 1000 cells are exposed to the absorption 

 of poison, and that under tlie condition of the experiment at least 



■XciOi ^^ TTTT "^'^ ^''^' molecules present are absorbed, then it would 

 require at least 10^ (one thousand million) molecules of mercuric 

 chloride to poison a single pulp cell of an apple such as Yates' 

 Pippin, whicli is one of the most resistant varieties. It is possible that 

 a lesser number of molecules, say, 10 to 100 millions, might be able to 

 arrest diastatic activity in a pulp coll without necessarily killing it. 

 To obtain some idea as to the intiuciice of diffusion and of convec- 

 tion currents upon the conveyance of poison, i.(>nq)ai ative experi- 

 ments were performed Avith applo* on water, and on melted 10 per 

 cent, gelatine allowed to set after adding poison. A preliminary 

 test showed that ordinary gelatine contains traces of poisonous 

 materials, but when well washed these are reduced to a mere trace. 

 The apples must of course be clean, and the gelatine sterilised by two 

 steam heatings, since fungi are able to develop on solutions 

 poisonous tf) apples. 



