538 



H. A. NOYES 



rate of one pound per minute, the door being opened thirty-five 

 minutes after it was first closed. 



Test 1. Two 8 ounce saltmouth bottles containing 99 grams 

 of water and two containing 90 grams of water. 



All four bottles were plugged with absorbent cotton. 



The results of this test are given in table 1. 



TABLE 1 

 Loss of water from dilution bottles in autoclaving 



Test 2. Fourteen 8 ounce saltmouth bottles containing 90 

 grams of water. 



Seven were plugged with absorbent cotton and seven were left 

 unplugged. They were set in the autoclave in sets of two. The 

 two at the rear were numbered 1 and 2, and the two nearest the 

 door 13 and 14. Even numbers denote bottles having no plugs. 

 Experiment conducted as in test 1. Results are given in table 2. 



It is evident from these tests that the errors due to loss of 

 water in autoclaving are considerable even with plugs in place. 

 When a number of successive dilutions are made these errors 

 would accumulate so that high dilutions might vary as much as 

 20 per cent from the dilutions desired. 



On the basis of these tests it might seem advisable to ascer- 

 tain by tests the amount of water that is regularly lost from 

 bottles of water sterilized in the auto-clave under specific condi- 

 tions. In some cases the loss might be constant enough so that a 

 fixed amount of water in excess of that desired could be added to 

 allow for losses occurring during auto-claving. 



