224 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



vitality of the seed itself certain special and more detailed 

 experiments were performed with Peas. 



In these experiments as indeed in all previous ones also 

 whenever the percentage of seeds which are still capable 

 of germinating reaches a low figure, below 10 p.c. for 

 example, it is necessary to plant a considerable number in 

 each experimental trial. With the smaller seeds about 

 300, with the larger about 150 are sufficient to ensure 

 accurate results. Using smaller numbers and especially 

 in the later trials contradictory or irregular results are 

 obtained. Thus if in a batch of 50 seeds only 1 p.c. are 

 capable of germination the result given on planting will 

 be either 2 p.c. or none, both of which are wrong. 



Bearing these facts in mind, and taking the proper 

 precautions to avoid errors arising from this source, 

 experiments were performed on Peas by placing some in 

 absolute alcohol and others in a saturated solution of 

 mercuric chloride in absolute alcohol. This salt is the 

 most deadly one which could possibly be employed and 

 its presence in the alcohol enables the penetration or non- 

 penetration of the latter to be ascertained in two ways, 

 firstly by comparing the number of seeds which germinate 

 after similar periods of immersion in either fluid and 

 secondly by testing the peas with ammonium sulphide. 

 This reagent throws down a black precipitate of sulphide 

 of mercury if a soluble salt of that base be present. The 

 Peas are separated into two halves, cutting the embryo 

 longitudinally and if the alcoholic solution of the mercuric 

 salt has penetrated the seed a black rim will form around 

 the margin of the cotyledons and on the outer surface of 

 the embryo on testing with ammonium sulphide. The 

 depth which the mercuric chloride has penetrated shews 

 that the alcohol has penetrated as far, if not further. For 

 since the salt is dissolved in absolute alcohol and since 



