THE SPINE-TAILED SWIFT. 



211 



On the following morning I found it hanging on the roughly-masoned sandstone wall. 1 

 hurried for my camera and exposed two plates through the door of the aviary to get the accompanying 

 side-view ; then in all haste 1 cut a hole in the front wire and obtained the back-view here reproduced. 

 There was no need for such haste, however, as the bird remained in this position without moving 

 for the next two hours, and on several occasions 1 again found it hanging on the wall. It still took 

 food and water, but nevertheless died alter a week's confinement. 



The photographs i learly show how the tail is used to support the body when hanging to a wall. 

 The body does not lie flat against the wall, for though the neck and the chest do so, the lower parts 

 are forced away by the thawing up of the tail to obtain support. The eight tail feathers, which are 

 spread, hardly touch the wall except at the very end. The sharply-pointed spines of each do all the 

 Supporting, and the two middle feathers arc stronger and the spines much heavier than those of the 

 remaining six. 



I should mention that my friend, Mr. Thos. Steel, F.L.S., kindly preserved this bird for me by 

 first soaking it in a dilute formalin solution and then drying it. This method of preserving natural 

 history specimens is well worth consideration, since in this case the appearance of the bird is exactly 

 as it was when it died.* 



LIFE HISTORY OF THE FISH, GALAXIAS ATTENUATUS. 

 By W. J. PHILLIPPS, Dominion Museum, Wellington. 



From time to time much has been written, and a great deal of discussion has taken place regarding 

 the life history of G ataxias attenuates, the Jolly-tail or Eel-gudgeon of Australia, and the Minnow 

 of New Zealand. Even so recently .is nn6 Meek (p 147) has not been willing to admit the statements 

 of various authors that the species spawns in the sea. Referring to a paper by McCulloch (1915, pp. 

 47-49), he states that the results of that author's investigations are not conclusive; and considers 

 it is evident that the young of about the size figured by McCulloch migrate to the sea, at which time 

 they are denatant. 



Hutton (1S72, p. 60) was the first to note that G. attenuates spawned in the sea, but others 

 interested in the subject were dubious about accepting his statement, since it relegated the species 

 into a class of its own, differing from all others of the genus by making a catadromus migration. 



Perhaps the most interesting observations were made by Clarke, 1899, PP- 78-79. He says : — 

 " Galaxias attenuatus periodically descends to the sea in January, February and March, where it 

 spawns, returning in March, April and May. The young begin to make their appearance in the rivers 

 sometimes as early as the end of June, but they definitely commence to arrive in August, the shoals 

 increasing in size and number in September and October. At the time of the advent of the fry I 

 have frequently, and at several places, seen large shoals of the ' inanga ' at sea, and have caught 

 specimens in verification, and have constantly observed them washed up by the breakers on to the 

 beaches near the mouths of large rivers, evidently when skirting the coast to enter them." 



Regan (1914, p. 41) has given the following interesting reference to this species and its allies : — 

 " Much has been made of the distribution of the Galaxiidae and Haplochitonidae, for some time regarded 

 as freshwater fishes found in southern Australia and Tasmania, New Zealand, and the southern part 

 of America. It is now known that Galaxias attenuatus, the only species common to all these regions, 



* See Aust. Naturalist, vol. iv, pt. 7, p. 99 (1919). 



