living Specimen of Siren lacertina. ^W 



watching, and then made a sudden dart upon the prey. Still, 

 however, he did not care for food oftener than once in a week 

 or ten days. When touched, he now changed his place with a 

 jerking motion, causing the water to spurt. '"^ 



Although I certainly would not have made the experiment 

 of the fragility of the siren, by throwing it on the ground'? 

 and although I would have hesitated to keep the animal out 

 of the water for several hours, while I knew that respectabl6 

 naturalists doubted if it would live more than a few minuted 

 out of that element, yet it so happened, that the animal, on one 

 occasion, made, of his own accord, an experiment (if it may 

 be so called) illustrative of both points. This was on the 18th 

 of May 182G (for the fact was recorded in my adversaria at the 

 time), soon after he had begun to be active and to take food foi' 

 the season. The water-box itself was ten inches deep : it was 

 placed on a plant-trellis or shelf, close by the lower end of the 

 sloping roof-sash of the greenhouse, and thus stood nearly three 

 feet from the ground. At tlmt period the box happened to leak; 

 and the gardener therefore filled it up with water between se- 

 ven and eight o^clock in the evening, at which time the siren was 

 seen safely lodged in the box. The door of the greenhouse was 

 locked, as usual, over night, and before it was opened in the 

 morning, the siren, to the great surprise of the gardener, was 

 found lying on a foot-path which passes round the exterior of 

 the greenhouse. I was speedily apprised of the circumstance; 

 and, on examining the spot, we could most distinctly trace, by 

 a shining glaze derived from his skin, the passage of the animal^ 

 through an edging of hcalli (Erica herbaceaj, and across a nar- 

 row flower-border, to a hole which he had scoojxjd out under the 

 brick-wall of the greenhouse, in escaping from within. The 

 foundation of this wall, it may be remarked, had intentionally 

 been made shallow or near to the surface, for the pur}K>sc of per- 

 mitting the roots of some shrubs, planted in the conservatory 

 style within, to }K»netrate to the exterior border. 



We jwssess no data for fixing witli certainty tlie nunil>er of 

 liours during which the animal had been out of the water. The' 

 box, as alreatiy mentioncil, being leaky, was filled near to the 

 brim iK^tween 7 and 8 in the evening : it Hcems likely that this 

 filling up had disturbed the animal, and that it had been ena- 

 bled partly to crawl and partly to glide over the margin, while 



