and Fish kept in a limited portion of Water. 279 



all find that will answer liis purpose. But I must observe that their 

 i^pecific gravity is continually tested : thus, having found what ap- 

 pears a suitable fibre, it is carried a little way, then projected to a 

 short distance from his mouth and watched as it falls ; if it falls 

 rapidly, it is again seized and carried direct to the nest ; if more 

 slowly, it is tried again in the same manner ; and if it then proves 

 too light, it is abandoned altogether and another selected. If a 

 piece is found better fitted for his structure than what he has 

 already obtained, it is rapidly conveyed to the spot ; much altera- 

 tion in the arrangement of the materials takes place, so as appa- 

 rently to dispose of the new prize to the best advantage, and it 

 is only after continued and indefatigable perseverance that he 

 succeeds in rearranging them to his wishes. If there should be 

 any strong fibre which he has a difficulty in causing to remain 

 in the position he requires, a small quantity of sand is brought 

 in his mouth and adroitly placed on the top of it to keep it down ; 

 if this does not effect the purpose desired so as to please him, 

 the refractory piece is taken out and rejected altogether. At 

 times he hangs or hovers close over the surface of the nest and 

 throws his whole body into a curious and rapid vibratory motion, 

 by which he causes a rapid current of water to be projected on 

 the materials as though it were to prove their stability ; and when 

 this operation is performed, the lighter particles and light mud 

 are as it were fanned or winnowed out by the generated current, 

 and may be seen floating away : this operation will also explain 

 the reason for testing the gravity of the materials before they 

 are used. Another very curious operation is the action of draw- 

 ing his body slowly over the surface of the materials which form 

 the nest. 1 believe that at this time he excretes a glutinous 

 matter which acts as a species of cement, and tends to keep the 

 materials together, at the same time that the pressure of his body 

 may render them more compact. Or it may be that the whole 

 surface of the nest is by this action charged with the milt, and thus 

 the impregnation of the eggs more perfectly ensured, as precisely 

 the same motion is employed after the eggs are deposited, and 

 from the appearance of the fish it seems to be attended with 

 pleasurable sensations. These two last-described operations are 

 very frequently repeated. 



If during this time any other male fish makes his appearance, 

 he is chased with the utmost ferocity and driven to conceal 

 himself in any cranny which he can find ; should, however, 

 another fish be also building, desperate battles ensue whenever 

 they approach each other^s position, or chance to meet while 

 collecting their materials. 



The whole time occupied in accumulating these materials for 

 the nest was about four hours, during which interval a goodly 



antity had been obtained ; and a small opening appeared to be 



