is 95 . ORIGIN OF SPECIES AMONG FLAT-FISHES. 173 



which enable us to some extent to form a judgment on the utility of 

 the generic and specific characters respectively. Z. punctafus and 

 Z. unimaculatns have been watched in the living healthy condition by 

 the late Mr. George Brook and by myself, and both of them found to 

 have the habit of adhering to the sides or glass front of the tank in 

 which the fish were confined. Z. norvegicus has not been observed in 

 the living condition. Mr. Brook in his first account of the habit in 

 Z. nnimaculahis stated that when the fish adhered to the glass front of 

 a tank the body was raised up from the surface, and the outer parts 

 of the fins tightly pressed against the glass. A constant current of 

 water passed out from the branchial chamber on the lower side, 

 between the body and the glass, and out beneath the hinder portion 

 of the longitudinal fins. This current was produced by a " vibratory 

 motion of the accessory portions of these fins." Brook's conclusion 

 was that the accessory portions of the unpaired fins were specially 

 constructed to aid in the respiratory function. In a later paper 

 Brook abandoned this view, stating that the motion of the 

 fins by which the current was produced was more vigorous in 

 the rays immediately in front of the tail than in the accessory flaps 

 situated beneath it. In a specimen of Z. punctatus observed by me, 

 the anterior parts of the fins were kept quite still and tightly pressed 

 against the glass. The lower surface of the tail and of the accessory 

 flaps of the fins were also in close contact with the glass. The other 

 parts of the surface of the body were separate from the glass. The 

 motion of the broadest part of the fins anterior to the tail was 

 undulating, and aided to cause the current of water which entered the 

 space between the body and the glass, not only by passing through 

 the mouth and gill-clefts, but also behind the operculum between the 

 lower jaw and the pelvic fins. The question arises, what causes the 

 fish to adhere to the glass ? Other flat-fishes adhere to smooth 

 surfaces for a time, but none keep in that position at rest for a long 

 time as the topknots do. Is it simply the adhesiveness of the 

 mucus on the epidermis ? Or is it that the pumping action of 

 the posterior fins causes a negative pressure beneath the body, 

 and so there is a slight pressure" on the outside of the fish ? There 

 is no sucker action of the ordinary kind, because the space 

 beneath the body is freely open in front to the outside water. 

 It was suggested by one of us who were watching the fish that the 

 tail fin and the accessory flaps together formed a small complete 

 sucker by themselves. The accessory parts of the unpaired fins were 

 quite motionless. I cut off these accessory portions with two snips 

 of a pair of scissors and returned the fish to the tank. The fish did 

 not alter its behaviour in the least ; it stuck to the glass when placed 

 there as easily as before, and seemed in nowise the worse for the 

 operation. The accessory flaps, therefore, do not appear to be 

 essential either to the respiration of the fish or to its adhesion to a 

 vertical surface. 



