3IO TYPES OF ANIMAL LIFE 



assumed a horizontal position, the body, when resting on 

 the ground, being supported b}^ the under surface of the 

 tail fin, and it may be that the posture assumed by the 

 Westminster specimen was due to one of its paddles 

 having been injured. They eat by preference lettuces 

 and endives, and these were always swallowed under 

 water, and they never eat when removed from it, tho\igh 

 food was repeatedly then offered them. When out of 

 the water they seemed to be oppressed with their own 

 bulk, and could only progress a few inches by means of 

 pressing their jaws and tail fin closely to the ground, and 

 making violent lateral efforts of the body, slightly sup- 

 ported by the paddles. 



The male devoured his food more rapidly than the 

 female, and thus obtained an undue share, so that it was 

 thought advisable to separate them at feeding time. 

 For this purpose a wooden partition, fitting into a 

 groove in- tJie floor and fastened by upright supports, 

 was occasionally let down into the tank, projecting a 

 few inches above the surface of the water. The female 

 took no notice of this alteration, but invariably waited 

 before commencing to feed until her mate was supplied 

 on his side with a portion. The necessity for the separa- 

 tion soon became apparent ; for the male cleared up 

 every scrap of food long before the female, a more dainty 

 and delicate feeder, had finished. He then became very 

 restive, swimming actively around his straitened quar- 

 ters, pressed his nose against the partition, rolled over 

 on his back, and exerted considerable force in his obsti- 

 nate and repeated attempts to remove the obnoxious 

 obstacle. Failing in his endeavours to push it on one 

 side, he next tried to get over it, lifted his head above 

 the w^ater, feeling the edge of the partition with his fore 

 paddles and raising them till they were almost level with 

 the projecting edge. 



