466 Simon Hemy Gage 



follows tliat the current going in at the mouth must be con- 

 stant, unless the respiratory movements are exceedingly slow. 

 The reason why the cephalically moving velum does not force 

 the water out of the mouth in narrowing the space between 

 the sieve and the velum is that, as the velum moves forward, 

 it leaves an equal space behind it and thus aspiration is pro- 

 duced in the branchial chamber, and as there is nothing to 

 support the thin mesal edges of the velar folds, the)^ move 

 laterally and thus make a free passage for the water to the 

 branchial chamber, so that the action of the velum alone tends 

 constantly to aspirate the water into the mouth. When the 

 velum is aided by the expanding branchial chamber in in- 

 spiration, an increased aspiration is insured and so much the 

 more is there a constant inflowing current. 



By careful experiment on transforming larvae it was found 

 that they continued to take a constant current into the mouth 

 even after they were able to attach themselves to the sides of 

 the vessel containing them by the almost completely devel- 

 oped sucking disc. Those experimented upon buried them- 

 selves in the sand and gravel whenever they were given op- 

 portunity ; it is believed therefore, that until the young lamp- 

 rey is entirely transformed and swims freely in the water or 

 becomes attached to a fish, water is inspired through the 

 mouth as well as through the branchiopores, but, in expiration, 

 it passes out only through the branchiopores, except when the 

 branchial apparatus is being cleared of particles taken in with 

 the respiratory currents. The action of the velum may be 

 most perfectly shown by thoroughly etherizing a larva and 

 then carefully removing the ventral body wall between the 

 velum and tentacles so that the velum may be very clearly 

 seen. If now the animal is set up endwise in water one can 

 study very satisfactorily the action of the velar folds. By ad- 

 ding ether occasionally one can control the rapidity of the re- 

 spiratory movements so that they may be slow enough for 

 careful study. For some purposes one may advantageously 

 remove the entire head cephalad of the velum. 



On a frontal section at the level of the branchiopores like 

 the one shown in PI. VIII, fig. 52, one can very readily see 



