THE ANCESTRY OF THE VERTEBRATES 639 



mostly, if not exclusively, derived from the dorsal ganglia. Hence the 

 possibility should not be denied that the ganglia of annelids and chordates 

 are homologous. 



The fact remains, however, that the chordate basis for a rellex action 

 is the spinal cord and not the spinal ganglia. Consequently, the accept- 

 ance of the Delsman hypothesis compels us to assume that the reflex 

 system of invertebrates has been lost, and a new reflex system invented 

 when annelids became chordates. The seriousness of this difficulty 

 seems not to have been fully appreciated by Delsman. 



With Hatschek (1878), Delsman suggests that both the dorsal longi- 

 tudinal musculature and the circular muscles of annelids have disappeared 

 in the metamorphosis of the annelid into the chordate. But he thinks 

 that the transverse muscle of amphioxus may correspond to the circular 

 muscle of annelids, although their position is not similar. 



The double muscular innervation, sensory and motor, demonstrated 

 by Boeke (191 1) in vertebrates is also found in annelids (Fraipont 1884). 



The absence of an ectodermal foregut in vertebrates, Delsman takes 

 to accord with expectation, if his theory is a correct one. 



Delsman derives the lateral eyes of vertebrates from the pit-eyes 

 of the preoral lobe of annelids. This conclusion, he thinks, is supported 

 by the embryological evidence that vertebrate eyes appear as pigmented 

 depressions of the neural plate in the region of the forebrain previous 

 to its closure. He assumes with Balfour (1881) that the forebrain of 

 vertebrates originates from the preoral lobe of annelids and amphioxus. 

 The preoral lobe or episphere of the trochophore larva contains paired 

 pigment spots which Delsman thinks can be traced into the paired eyes 

 of vertebrates. A step in the vertebrate direction is taken by those 

 annelids in which the pit-eyes become vesicular eyes. The absence of 

 eyes in amphioxus is interpreted, in agreement with Ray Lankester (1880) 

 and many others, as a result of degeneration. Delsman holds that verte- 

 brate eyes became secondarily included within the brain and, as the 

 animal became more opaque, grew out towards the skin. If, then, 

 the prechordal part of the vertebrate embryo is the homolog of the 

 preoral lobe of amphioxus, the prechordal forebrain and the eyes, would, 

 according to Delsman, be derived from the same source which in anneUds 

 gives rise to the cerebral ganglia and the eyes. The otic vesicles of some 

 annelids are paired organs lying just behind the mouth, in a position which 

 corresponds to that of the otic vesicles of vertebrate embryos. 



Hatschek (1909) defended the assumption that the infundibulum 

 of craniotes represents the anterior boundary of the acraniote brain. 

 But Hatschek thought that the prechordal portion of the vertebrate 

 brain is an outgrowth from the epichordal region. This is not Delsman's 

 view. Delsman assumes a forward extension and incorporation of ecto- 



