CUAP. IV.] 



CLIMBING FISH. 



215 



those with labyrinthiform pharyngeal bones, so disposed 

 in plates and cells as to retain a supply of moisture, 

 which, whilst crawling on land, gradually exudes so as 

 to keep the gills damp. 1 



The individual which is most frequently seen in these 

 excursions in Ceylon is a perch caUed by the Singhalese 

 Kavaya or Kawhy-ya, and by the Tamils Pannei-eri, or 

 Sennal. It is closely allied to, if not identical with, the 

 Anabas scandens of Cuvier, the Perca scandens of Daldorf. 

 It grows to about six inches in length, the head round 

 and covered with scales, and the edges of the gill-covers 

 strongly denticulated. Aided by the apparatus already 

 adverted to in its head, this little creature issues boldly 

 from its native pools and addresses itself to its toil- 

 some march generally at night or in the early morning, 

 whilst the grass is still damp with the dew; but in its 

 distress it is sometimes compelled to travel by day, and 

 Mr. E. L. Layard on one occasion encountered a number 

 of them travelling along a hot and dusty gravel road under 

 the midday sun. 2 



Nat. dcs Paisson*, torn. vii. p. 246. 



2 Annah and Mag. of Nat. Hist., 

 May, 1853, p. 390. 'Mr. Morris, the 

 government-agent of Trincomalie, 

 writing to me on this subject in 

 1856, says "I was lately on duty 

 inspecting the bund of a large tank 

 at Nade-cadua, which, being out of 

 repair, the remaining water was con- 

 fined in a small hollow in the other- 

 wise dry bed. "Whilst there heavy 

 rain came on, and, as we stood on the 

 high ground, we observed a pelican 

 on the margin of the shallow pool 

 gorging himself; our people went 

 towards him and raised a cry of fish ! 

 fish ! We hurried down, and found 

 numbers of fish struggling upwards 

 through the grass in the rills formed 

 by the trickling of the rain. There 

 was scarcely water enough to cover 

 them, but nevertheless they made 

 rapid progress up the bank, on which 



our followers collected about two 

 bushels of them at a distance of forty 

 yards from the tank. They were 

 forcing their way up the knoll, and, 

 had they not been intercepted first 

 j by the pelican and afterwards by 

 j ourselves, they would in a few 

 minutes have gained the highest 

 point and descended on the other 

 side into a pool which formed another 

 portion of the tank. They were 

 chub, the same as are found" in the 

 mud after the tanks dry up." In a 

 subsequent communication in July, 

 1857, the same gentleman says 

 " As the tanks dry up the fish congre- 

 gate in the little pools till at last you 

 find them in thousands in the moistest 

 parts of the beds, rolling in the blue 

 mud which is at that time about the 

 consistence of thick gruel." 



" As the moisture further evapo- 

 rates the surface fish are left un- 

 covered, and they crawl away in 

 4 



