appearance and habits 3°9 



Association with Nomeus 

 The association with Physalia of the little purplish-barred stomateoid, the rudder-fish, Nomeus 

 gronovii is interesting. It has long been known to shelter under Physalia, although I never saw one in 

 the Canaries, and it was thought to lure other fishes into the reach of the tentacles; it was also con- 

 sidered to be immune from their stings. But Garman (1896) reported partially digested Nomeus 

 attached to the tentacles, and an 18 cm. long specimen of a male Physalia taken by 'Discovery II' 

 between stations when testing gear, on 29 December 1929, in a position 3 22' S., 32 25' W., is 

 accompanied by a label in the handwriting of Dr Stanley Kemp, recording that 'twenty or more small 

 Nomeus were seen swimming below the Physalia and in one dip the latter was caught as well as several 

 of the fish. In the net the fish came into contact with the tentacles of the siphonophore and were 

 immediately killed.' Kojiro Kato (1933) observed that Nomeus vigorously attacked Physalia from 

 below, eating parts of it, including the tentacles. On opening the stomach of one fish as a check, he 

 found unmistakable Physalia tissues inside. 



I am indebted to Mr J. C. Natzio for a note on another stomateoid fish-associate. Natzio, a senior 

 boy of Charterhouse School, sailed aboard R.R.S. ' Discovery II ' to assist in the biological work of the 

 spring cruise of 1958, and referred to the fish as Lirus maculatus. I understand that it is better known 

 as Schedophilus maculatus (Gunther), but I have found no reference to this association in the Physalia 

 literature. On a few days following 17 March, when the ship was frequently stopped on station, the 

 fishes were often seen to swim around and beneath tentacles of medium sized Physalia, and both 

 animals were often taken together in hand-nets. On several occasions Schedophilus was noticed to 

 swim from one Physalia to another. In an aquarium one of the fishes escaped after being entangled 

 for about 10 min., and continued to swim around with apparently no ill effects. Mr Foxton, who was 

 also on board, told me that another specimen of Schedophilus after being stung by Physalia towed the 

 Physalia along the surface of a tank, later escaped, rolled on its side, but eventually appeared to 

 recover. This fish, like Nomeus, is banded with blue, though the colour is not very conspicuous. 



Rolling- or Somersaulting-Behaviour (PI. X, figs. 1-6) 

 Much time was mistakenly spent at Arrecife watching rolling-behaviour, filming it and taking still 

 photographs in the laboratory. It was only much later on, when the specimens were no longer avail- 

 able, that I realized that rolling takes place much more frequently in the still air of the laboratory than 

 at sea in a good breeze. My tentative conclusion is that Physalia, when in a state of muscular tonus 

 with crest erect, is in unstable equilibrium and is unable to remain upright without the aid of the 

 force exerted by the wind. In the relaxed position, assumed after some time in the laboratory, it does 

 remain upright but with the crest deflated, and the overall length increases, and the width decreases. 

 In a note on sailing-experiment 3 (see page 318) I mention that specimen 1 fell over into the wind when 

 the breeze dropped. 



The older naturalists interpreted rolling-movements as change of 'tack', but the reason for it 

 seems to be as follows : in the contracted, inflated condition with crest erect, Physalia is asymmetrical 

 because most of the large tentacles, gastrozooids and gonodendra lie well out on the windward side, so 

 that the heeling force thus set up prevents the animal from capsizing in a breeze. When the wind 

 drops, the animal falls into the wind, and complicated muscular contortions are then needed to get 

 back into the sailing position. This of course needs confirmation under conditions open to the wind, 

 but I have certainly observed at sea in winds of force 3 or 4, a number of animals steadily drifting 

 along without rolling; on the other hand I have not had an opportunity to watch them in a calm except 

 under laboratory conditions. It is significant that Huxley (1859) in commenting on this somersaulting- 



