1 6 Mai' shall Ward and Dunlop. 
0 i. To this was added some of the solution A (i.e. aqueous 
extract of endocarps 1 . No precipitate was formed — the 
solution remained perfectly clear for two days. 
Oi. To this were added a few drops of solution K (i.e. 
glycerine extract of seeds). Remained perfectly clear for 
several hours, but a precipitate fell during the night 2 . 
P. I then took the test-tube E, and filtered the clear yellow 
liquor off from the pericarps, and divided the solution into 
three parts, marking the test-tubes P i, A 2, P 3 respectively. 
(1) Pi. Added a few drops of the solution G (aqueous 
extract of seeds) : an abundant precipitate fell during the night. 
(2) A 2. Added a few drops of solution / (glycerine extract of 
endocarp) : no trace of turbidity or precipitate was observable 
next day— nor after forty hours. 
(3) A3. Added a few drops of solution K (i.e. glycerine 
extract of seeds) : slight precipitate fell during the night and 
increased slowly during the following day. 
Series III. 
1 now prepared a series of solutions of different parts of the 
seeds, each in a labelled test-tube, and placed them in the 
hot-house for the night. 
No. a. The testa only of the seeds, with distilled water. 
(The liquid remained quite transparent and clear for two 
days, and then became turbid as Bacteria developed.) 
No. b. The testa only used, and pure glycerine poured on it. 
A perfectly clear, colourless solution resulted. 
No. c. Endosperm and embryo cleaned of testa, taken and 
extracted with cold distilled water. 
A limpid colourless extract resulted. (Traces of turbidity 
next day, and after forty-eight hours it was dirty-white and 
cloudy 3 .) 
1 I only repeated this experiment here to make the series more complete : no 
trace of the ferment occurs in the endocarp. 
2 The precipitate formed much more slowly than when the aqueous extract of 
the seeds were used. 
3 Bacteria, &c. 
