GERMINATION OF SEEDS. gfjf 



mer situation. After continuing this practice for some time, Air of : 

 while engaged in the same manner, 1 was hastily called ™ffi^£ 

 away, and left the quill partly in the jar, with one end rising mercury ty* 

 out of the mercury. '1 he jar was then two thirds full bf fe * th€t » 

 gas, but on my return in about half an hour, I perceived it 

 had increased very considerably, and on placing the quill 

 in other jars, I distinctly heard a shrill whistling noise, like 

 that of air under pressure passing through a capillary tube, 

 and I observed the mercury slowly to sink, till it was on a 

 level on the inside and outride of the jar. I was then con- 

 vinced the atmospheric ai'* had rushed in by means of the 

 quill, and consequently that all the experiments, in which 

 this had been introduced, must have been vitiated. I re* 

 versed the quill, and it still had the same effect. I tried it but rrot 



in jars over water, but no air passed. I afterward made use though****. 



J l String, &<vth«r 



of string, and other substances, and they all admitted the substances*^*- 



air through quicksilver, though in different degrees, some edmthesMiW 



being much slower conductors than others. After consider- 



ing this phenomenon, the best judgment I am able to form 



of it is, that the air does not pass through the body of the 



quill, or other substance, but between the mercury and its 



sides ; and in water the passage is prevented by their being Mercury does 



in closer contact with each other. Whether this explanation notformaclose 



contact, 



be satisfactory, I leave to your superior knowledge to deter- 

 mine : 1 confess I was gratified with the discovery, as far as 

 concerned my experiments, ns it enabled me to prevent their 

 being so rendered incorrect for the future. 



It having been stated as a principal argument in favour Argument for 

 of the emission of solid carbon from the seed to unite with the emission of 

 the oxigen gas of the air, that the quantity of carbonic ueid termination ** 

 produced was found to be equal to that of the oxigen gas questioned* 

 disappearing; upon reflection, it appeared to me replete 

 with difficulty, if not impossible, to ascertain this to any 

 degree of aecuracy, from the moistened seeds never ceasing 

 to give out carbonic aeid gas, whether oxigen gas be pre- 

 sent or not. 1 was therefore desirous of informing myself 

 upon this subject, and for this purpose 1 instituted the fol- 

 lowing method of proceeding. 



Exp. 1. into an inverted jar, containing about 13 cubic Experiment 

 iache* carefully filled with mercury, I introduced a consi- to ascertain tht 



derable ac * 



