BREATHING MECHANISM OF THE LAMPREYS. 93 



enough to reach from the mouth opening to a point about I cm. 

 back of the semiannularis muscle and establish a connection 

 between the mouth cavity and that of the pharynx. At first the 

 animal made fruitless attempts to attach itself and finally did so 

 as firmly as ever. Upon examination, the animal was found to 

 have swallowed the tube. A tube of the same length, but having 

 a hook at one end, was now inserted into the mouth as before. 

 The hooked end lay between the tongue lobes but did not inter- 

 fere with the oral funnel in any way. It was found that the 

 lamprey was neither able to swallow the tube nor to attach itself 

 but lay on its side. Finally it was able to attach itself very 

 feebly, but in order to do so the tongue worked back and 

 forward continually, thus creating a slight vacuum. 



(3) -A- guarded bristle was passed through the second external 

 gill opening of a living lamprey into the water tube and on 

 through the mouth so that the enlarged tip of the bristle lay 

 between the tongue lobes, but did not interfere with the action of 

 the funnel. The bristle thus prevented the complete closure of 

 the velar valves and also the close approximation of the lateral 

 tongue lobes. The animal was unable to attach itself except 

 feebly by the aid of the continuous movement of the tongue 

 already described. The bristle was drawn back of the semi- 

 annularis muscle so that it could no longer interfere with the 

 action of the tongue, but was left where it would interfere with 

 the action of the velar valves. The animal immediately regained 

 the power to attach itself as firmly as before, thus showing that 

 the velar valves do not aid in forming the vacuum in the mouth 

 cavity. 



From these experiments it follows that the partial vacuum 

 which effects the attachment of the lamprey exists in the mouth 

 and oral funnel only. The movement of the tongue at the mo- 

 ment when the animal attaches itself shows that the vacuum is 

 produced at least in part by the pumping action of the tongue. 

 The fact that after the animal has attached itself, the tongue pro- 

 jects forward, shows that the communication between the mouth 

 cavity and the pharynx is not kept closed by the tongue, but prob- 

 ably by the semiannularis muscle, thus leaving the tongue free 

 for other purposes. 



