TRANSACTIONS OP SECTION D. 915 



aUinentary system are formed into diverticula, the cells of which are metamor- 

 phosed iuto chordoid tissue closely resembling that of the notochord. These 

 structures, occurring as they do in forms which reveal more and more intimate 

 resemblances to the higher Chordata with the progress of research, have the same 

 relationship throughout life to the alimentary canal (pharynx) that the vertebrate 

 notochord has to the same organ in the young stages of the higher Chordata. 



For this reason they are accepted by some morphologists as organs homologous 

 to the notochord, but exemplifying a more primitive condition of the same. To 

 those who will not accept this conclusion their resemblance to the vertebrate 

 notochord must appear as a most remarkable instance of convergent evolution 

 occurring in animals which in many other respects show close genetic relationship. 

 If, however, we accept this view we are led to ask the question, Does the study 

 of these chordoid structures throw any further light upon the primary origin and 

 function of the notochord ? 



In Balanof/lossus the function of the so-called notochord is usually assumed to 

 be that of support to the proboscis and its muscles, so that it appears to have 

 already acquired that secondary function of support to the mesodermic tissues 

 which, in the higher Chordata, leads to its eventual loss of connection with its 

 parent tissue. 



In Cephalodixcus, however, the two pleurochords run as dorso-lateral chordoid 

 grooves throughout the length of the pharynx. Posteriorly they arise at the com- 

 mencement of the oesophagus, and anteriorly they curve round to left and right to 

 open to the exterior by two apertures, one on each side of the mouth. These 

 apertures have been identified as pharyngeal clefts, and will be referred to again 

 later. The actual function of the pleurochords in Cephalodisciis is, by the nature 

 of the case, incapable of demonstration, but the further structure of the pharynx gives 

 a clear indication of their use. 



In Ammocwtes, in Amphio.xus, and in the Tunicatu a system of glands and 

 ciliated grooves, comprising in its full development asubneural gland, an endostyle, 

 a pevi-pharyngeal band and hypobranchial groove, has been demonstrated. This 

 system varies in its lesser details in the types mentioned, but its constant occur- 

 rence indicates clearly that the gnathostomatous condition of the Vertebrata wms 

 preceded by a method of feeding which depended on the ingestion of microscopic 

 food suspended in water cttrrents and the subsequent separation of the former from 

 the latter. A similar system can be demonstrated in Cephalodisciis, the organ for- 

 merly described as a notochord being the subueural gland, connected by a well- 

 defined peripharyngeal groove with the ventral alimentary portion of the gut, which 

 itself can, through Balanoglossicit, be homologised with the endostjde. The whole 

 structure of the pharynx points to the conclusion that, whilst these organs serve to 

 collect the food particles and carry them through the oesopiiagus to the stomach, 

 the pleurochords serve to conduct the water current forwards and eventually out- 

 wards by the pharyngeal clefts. In other words, the notochord of the vertebrates 

 has arisen from the endoderm, as a certain specialised area of the alimentai-y canal, 

 which, becoming stiffened by a chordoid metamorphosis, serves as an organ for the 

 removal of the water current involved in the ciliary ingestive processes. 



According to this hypothesis the hypoblastic origin of the notochord, no longer 

 a difficulty, becomes a phyletic repetition of the same nature as the epiblastic origin 

 of the nervous system. 



In the case of the pharyngeal clefts Dr. Harmer has already pointed out that 

 in Cephalodiscus they serve, in all probability, for the discharge of ' atrial ' water, 

 and it has been shown above that they are morphologically merely the openings of 

 the pleurochords to the exterior. In this species there is no indication that they 

 function for respiration. They are kept open by chordoid walls, and have the 

 same relation to the pleurochords as has the anus to the alimentary canal. The 

 hypoblastic origin of a gill slit is thus explained in the same way as that of the 

 notochord, and the primary origin of the two has a similarity almost amounting to 

 identity. The pharyngeal cleft, like the notochord, later on in the history of the 

 Chordata, loses its primary atrial function and becomes a branchial gill slit. Just 



3n2 



