for the float to initially resist the hydrostatic pressure. 

 The various metabolic functions of the colony, as well as 

 vigorous swimming, will necessarily place an additional 

 drain upon the available energy. Furthermore, as yet we 

 know nothing of the energy requirement of the enzymatic 

 machinery responsible for gas production. And, in addition, 

 the diffusive loss of carbon monoxide, which may be as high 

 as 0. 1 mm 3 per hour (table 3), will demand that gas produc- 

 tion be sustained at a maintenance level at least this high, 

 as indeed it was in Experiment 6. It thus appears likely, 

 on the basis of these observations, that a siphonophore 

 colony may require a period of several hours in which to 

 complete the process of filling its pneumatophore. If, how- 

 ever, instead of an average oxygen consumption, we employ 

 the elevated rate shown to be associated with gas secretion 

 (Experiment 6), the work required to counter hydrostatic 

 pressure in one hour becomes less than 25 percent of the 

 energy available from respiration (oxygen consumed). 



DISCUSSION AND CONCLUSIONS 



It is apparent from the foregoing that our knowledge 

 regarding the gas dynamics of N. bijuga is still fragmentary 

 and in some cases quite inadequate. Nevertheless, we are 

 still able to make some predictions concerning its role as 

 a sound scatterer. 



On the basis of volume measurements, it now seems 

 most likely that nanomians are comparatively sensitive to 

 internal gas pressures rising above ambient. There is little 

 evidence at present to suggest that the pneumatophores are 

 very elastic, and in any case it is unlikely that they could 

 tolerate an expansion equivalent to more than 10 to 20 per- 

 cent of the float volume. The structure of the float itself 

 indicates that even much smaller additional quantities of 

 gas are probably intolerable and must be voided. We are 

 faced, therefore, with a population of organisms each of 

 which must void gas during its diurnal vertical ascent. That 

 this is in fact the case has been suggested previously. 1 ' 1 

 It is not yet certain how many gas bubbles are likely to be 

 expelled by a rising nanomian, but an estimate of one per 

 atmosphere of reduced ambient pressure does not seem 

 unwarranted. Furthermore, some of these transient 



27 



