638 ECOLOGY AND BEHAVIOR 



temperature and survival times differed significantly over the two experi- 

 mental years reported in this study, yet the patterns of change observed in 

 serum glucose, hematocrit, cholesterol, etc., did not. 



3. If experiments are conducted on animals starved in live cars, the time 

 held in captivity, water temperature, and hematocrit should be recorded. 



In the present study, the survival of a shark was best predicted by con- 

 sidering alterations in hematocrit and serum proteins: 



Y = 0.267 X + 4.625Z - 5.71 r 2 = 0.74 



p = 0.0001 



where 



Y = days until death 

 X = hematocrit value 



Z = serum protein, gm% 



However, the change in hematocrit alone was a reasonable predictor of 

 survival time: 



Y = 0.43 X - 2.2 r 2 = 0.65 



p =0.001 



The slope of this relationship would presumably change under different 

 experimental conditions, but the hematocrit value at death (^5%) would still 

 be useful as an "endpoint" for calculating survival time once the rate of 

 change in hematocrit was experimentally determined. 



If hematocrits were determined on freshly caught animals and redeter- 

 mined for the same animals prior to their use as experimental subjects, a 

 stable hematocrit could be taken as a favorable indication, while a sharp 

 decline would indicate that physiological alterations were occurring. Since 

 hematocrit can be conveniently determined with a minimum of equipment, 

 it has a number of advantages as a diagnostic tool. 



SUMMARY 



1. Squalus acanthias were placed in live cars to assess the effects of 

 confinement and starvation on this species. 



2. Four experimental groups were used, and blood and tissue samples 

 were taken at intervals throughout a period of confinement of up to 23 days. 

 These were compared with samples from freshly caught animals and animals 

 left undisturbed in the experimental pens. 



3. Behavioral changes involved a gradual decrease in spontaneous 

 swimming and a loss of equilibrium followed by immobility and death. 

 Attempts to induce feeding were unsuccessful. 



4. Physiological alterations took place immediately on introduction to 

 experimental pens; they included increases in serum levels of sodium, 



