1914] DARSIE, ELLIOTT 6* P EI RCE— GERMINATING POWER 105 



the variations in the temperature were so slight that it was not 

 thought to be worth while to continue to record the readings. The 

 slight variations in room temperature shown in the record given 

 below are due mainly to the opening of the room, the presence of 

 one or more of the experimenters, and the heat liberated by a 32 

 candle-power incandescent light bulb. The light was turned on 

 only as needed, but it is obvious that if the temperature of the room 

 is moderately low, as it was, the heat liberated from a carbon fila- 

 ment lamp of considerable candle-power and radiated and exhaled 

 from the body of an adult individual of average stature, during 

 the 10-40 minutes required for work or observation, would be con- 

 siderable, and in a smaller room would make a noticeable change 

 in the temperature. In our case the fluctuations were slight. 



The thermometers were of two sorts, short and long, the one 

 requiring to be pulled part way out of the flasks to be read, the other 

 long enough to make this unnecessary. Both read to o?i C. The 

 differences in the thermometers, flasks, and cotton plugs used to 

 close the flasks and to hold the thermometers steadily in the necks, 

 as well as the inevitable differences among the seeds themselves, 

 are responsible for such lack as there may be of uniformity in the 

 corresponding results. 



The flasks were sterilized by being washed in corrosive sublimate 

 solutions, generally saturated aqueous, and then thoroughly washed 

 out with boiled and cooled distilled water. In many cases the 

 seeds could not be sterilized, with the methods which we employed, 

 without impairing their vitality. It is a matter of very consider- 

 able practical importance to find an agent which, at the same time 

 that it is inexpensive, will effectively sterilize the surface of seeds 

 without harming the germ within. We tried copper sulphate in 

 various concentrations, but found it unreliable and often injurious. 

 In most instances we used a saturated aqueous solution of corrosive 

 sublimate. This is thoroughly efficient wherever it penetrates, but 

 a considerable part of the surface of many, if not most, seeds is 

 covered with a film of air, hard to dislodge, which prevents the 

 sterilizing solution from reaching the spores or bacteria which 

 may be adhering to the seed coats. Schroder 7 has discussed the 



7 Schroder, EL, Die Wiederstandsfahigkeit des Weizen- und Gerstenkornes gegen 

 Giftc und ihre Bedeutung fur die Sterilization. Centralbl. f. Bakt. 28:492. 19*0. 



