MR GOODSIR ON THE ANATOMY OF AMPHIOXUS LANCEOLATUS. 259 



cimens, I agree with MULLEB in considering its existence as highly problematical, 

 and I shall proceed to demonstrate that, in accordance with the plan on which the 

 other organic systems of this animal are formed, such an opening into the branchial 

 chamber could not exist. The abdominal openings in the Myxine are the result of 

 the closure of its numerous branchial clefts by the integuments. They are analogous, 

 in fact, to the branchial orifices of the tadpole, immediately before cessation of the 

 aquatic respiration. The respiratory apparatus of the Myxine, then, although infe- 

 rior in functional activity to that of other fishes, is actually referable to a more ele- 

 vated type. The Myxine possesses a brain in which the central masses are consider- 

 ably evolved, and a nervus vagus of sufficient development. The brain of the Lance- 

 let, again, is reduced to a mere filament, and the existence of a nervus vagus appears 

 to be highly problematical. These considerations, and the fact that branchial 

 openings have not been detected by YARRELL, COUCH, MULLER, or myself, must 

 lead to the conclusion that this fish has either never had branchial clefts at any 

 period of its existence, or that if it at any time had them, they must have totally 

 disappeared. I am inclined to believe that the former is the real state of the case, 

 not only from the views already urged in reference to the other organs in this 

 animal, but also from the consideration that if these clefts had ever existed their 

 traces would have remained. As the seventy or eighty pairs of branchial ribs 

 cannot be looked upon as true branchial arches, and as we cannot suppose that 

 any vertebrated animal could have so many branchial fissures, we are driven to 

 the conclusion that the Lancelot never had at any period of its existence true 

 branchial arches and clefts, and that the ribs have been developed for a special 

 purpose for a mode of branchial respiration hitherto undescribed in the class of 

 fishes. 



The Lancelet respires by receiving sea- water into the anterior compartment 

 of its intestinal tube this cavity is kept dilated by the elasticity of the numerous 

 filamentous ribs, and this dilatation may be increased by the action of the super- 

 imposed ventral bundles of the lateral muscles. It is contracted by the action of 

 the abdominal muscle. This is a mode of respiration similar to that which pre- 

 vails in the tunicated mollusks. It is interesting to observe that the branchial 

 membrane of the Lancelet is exactly similar in its peculiar vascularity (ramifications 

 at right angles) to that which lines the branchial cavity of the mollusks just 

 specified. 



If the branchial membrane were examined in the living animal, it would un- 

 doubtedly exhibit cilia in as great abundance as in the branchial membrane of the 

 ascidice, and such a ciliary arrangement must constitute one of the active agencies, 

 not only in renewing the supply of water for respiration, but also in conveying 

 food to the orifice of the digestive portion of the intestinal tube. As in the ascidice, 

 the entrance of the intestino-respiratory canal is guarded by filaments. The hyoid 

 filaments of the Lancelet performing the same office as the filaments at the oral 



VOL. XV. PART I. 4 A 



