238 Merkel Henry Jacobs 



J Influence of the Conditions Attending Desiccation 

 a Effect of Rapidity of Drying 



Since the time of Spallanzani it has been the almost unanimous 

 testimony of all observers that rotifers dried on a clean glass slide, 

 even for short periods of time, are killed, while when a Httle sand 

 or moss is present this is not the case. These factors have been 

 variously explained. Spallanzani thought that the sand was neces- 

 sary to protect the animals from the injurious action of the air. 

 Ehrenberg and others that it held in its interstices sufficient water 

 to prevent actual desiccation and Davis that in its presence the 

 rotifers w^ere able to protect themselves by means of a waterproof 

 cyst. All of these views have been shown to be untenable. That 

 it is not contact with the air that kills the animals is shown by 

 Doyere's experiment in which rotifers and tardigrades dried in 

 the air showed a lower mortahty than those dried in a vacuum. 

 That the sand in itself cannot serve as a protection against evapo- 

 ration is shown by the fact that it may be heated to a temperature 

 of 100° C. or more without kiUing the rotifers contained in it. 

 That Davis' explanation is not the correct one is shown by the 

 fact, discussed at some length above, that even in rotifers dried 

 with sand a true desiccation occurs. We are, therefore, forced to 

 the conclusion that the favorable effect of the sand is due rather to 

 a retardation of desiccation than to its entire prevention. 



To test this view three lots of rotifers were taken from the same 

 culture and the first dried without any sand, the second with a 

 small quantity, and the third with a large quantity. All were then 

 kept at room temperature under exactly the same conditions for 

 four days, at the end of which time water was added and the roti- 

 fers kept under observation. In the following table the first 

 column gives the number of individuals experimented upon, the 

 second the average time required for movements to appear in a 

 given number and the last column the percentage that eventually 

 recovered. The number chosen for the second column was usu- 

 ally 10 per cent of the entire number of individuals and except 

 where otherwise mentioned the determinations were made on this 



