3i° 



DISCOVERY REPORTS 



behaviour said, ' . . . I have over and over again watched the operation going on with great vigour in a 



dead calm '. 



Floatation: Pneumatocyst or Air-sac 



I was surprised, when I first dissected a live Physalia by making a vertical cut with scissors down the 

 outer pneumatocodon wall, to see how quickly this muscular coat retracted and freed the pneumato- 

 cyst ; and particularly to see the branching digitiform processes of the latter which fit into the pockets 

 of the crest. Because little attention seems to have been paid to them, I give a photograph (PI. X, 

 fig. 7) of a fixed specimen. To obtain fixation of the extended air-sac, which is covered inside, except for 

 the area of the gas-gland, by a thin chitinous layer, it is necessary to maintain pressure for some time 

 on the sac while it is in the fixative. The raising and lowering of the crest evidently depends on 

 muscular compression and relaxation of the float and the air-sac, as described by Leuckart (1851). 

 Lesson (1843) gave a crude figure (M, pi. 11). 



Text-fig. 1. Physalia phy salts. A right-handed specimen, number 58, x | (Discovery St. 3255). Plan-view from 

 above to show the septa of the crest. I, II = cormidia I and II (main zone). 



The textbook account of Physalia by Delage and Herouard (1901) contains statements — no autho- 

 rity quoted — about crest-structure which need correction. First of all I have never seen any sign of an 

 internal longitudinal septum in the crest. Secondly there is no perforated septum joining the edges 

 of the primary and secondary septa, and partially enclosing the outer pockets of the pneumatocodon. 

 In consequence it is untrue to say that the finger-like processes of the saccus form horizontal pairs, one 

 on each side of a longitudinal septum. The authors also wrongly describe the gas-gland as strip- 

 shaped: it is nearly circular. Their diagram of the relationship between crest-pockets and saccus, 

 shown in their pi. 28, fig. 4, is useful. 



Here is a description of the crest of a specimen well preserved in an extended condition — specimen 

 no. 58, taken by R.R.S. 'Discovery II' at St. 3255 on 13 June 1955 (in 41 08' N., 14 32' W.), right- 

 handed, of floatlength 16 cm., gas-gland diameter 45 cm. and pneumatocyst-capacity about 300 c.c. 

 The first developed set of septa, ten in number, hang down some 3 cm. in the central region. The 

 major pockets are subdivided three times, and rudiments of a fourth series are visible at the crest- 

 margin. The secondary septa are 2-5 cm. deep, the tertiaries 1 cm. and the fourth series 0-5 cm. A 

 view of the float and crest from above is given in Text-fig. 1 . 



