R KiDD AND C West 9 



the net yield, the effect of temperature during the period of soaking is 

 most striking as is shown by the following table : 



Table X. Germination and growth results. 



Period of Temperature during Total No. of Actual dry weights of Dry weights 



soaking the soaking treatment germinations the plants produced per'lOO plants 



1 day WC. 9 ]-805g. 20 05 g. 



Iday 20° C. 12 2-525 2104 



1 day 25° C. 18 3-6(50 20-33 



Iday 30° C. 16 3 070 19-18 



Considering these results from a plant-physiological point of view, it 

 is curious to note that we do not obtain the simple result which might 

 have been expected, namely, that both number of germinations and 

 yield per plant should fall ofE, with increasing time of soaking, most 

 gradually at low temperatures and most rapidly at high temperatures ; 

 in other words, we might have expected the highest values for plant 

 weight and number of germinations at the lowest temperature for any 

 given time of soaking. In Table IX the highest figures obtained for any 

 given time of soaking are printed in heavy type and show at a glance 

 that this is not the case. 



Experiment 4. Seeds sown on damp sand. 



In Figs. 4 and 5 (Plate II) the results of further experiments with 

 another variety of Dwarf Bean are shown. In these cases a much longer 

 subsequent growth period {i.e. 8-10 weeks) was allowed. The difference 

 in the results from seeds sown dry (Figs. 4 a and 5 a) and seeds which 

 had been soaked for 12 hours previous to sowing (Figs. 4 b and 5 b) is 

 very marked. 



Removal of the testas before soaking apparently has no effect upon 

 the results of the soaking treatment (Figs. 5 a and 5 b). 



DISCUSSION AND SUMMARY. 

 Three main conclusions may be drawn from the results obtained from 

 the experiments described in this communication, although these were 

 not carried through to the final stages of development of the various 

 plants concerned. In the first place, soaking the seed in distilled water 

 previous to sowing may have a marked effect upon the subsequent 

 growth of the plant. In the second place, a germination test cannot be 

 relied upon in the least to give any criterion of what this eff'ect may be. 

 And thirdly, the nature of the effect is strongly specific, quite different 



