1897.] Cephalic Homologies. 935 
1. Is Amphioxus a Tunicate becoming a Vertebrate by the 
acquisition of segments? 
2. Is Amphioxus an Annelid becoming a Tunicate by the 
acquisition of a notochord and gill bearing pharynx ? 
Of course, we all understand that Amphioxus is probably 
not the exact transitional form, but merely the nearest living 
ally, known to us, of the transitional form. 
5. FURTHER AFFINITIES OF AMPHIOXUS. 
We have already maintained that the nearest affinities of 
Amphioxus are with the Tunicates. We now are confronted 
by the question: Are its affinities on the other side closer with 
A, the Vertebrates, or B, the Annelids? Here I believe we 
we must start with the decision to leave the pharynx and 
notochord out of consideration, because these structures are 
characteristic of the Chordata as a whole and not of the Ver- 
tebrata. But there are three sets of organs in Amphioxus which 
offer special peculiarities morphologically ; these are the seg- 
ments, the sexual organs and the excretory organs. Let us 
consider these in order. 
1. Segments—The mesothelial segments of Amphioxus arise 
with their cavities in open communication with the archente- 
ron. Ray Lankester has advanced, and O. Hertwig and 
others have adopted, the hypothesis that the coelom of Verte- 
brates is an enterocoele, but hitherto, although special investi- 
gations have been made, no conclusive evidence has been 
found that in any true Vertebrate there is a clear resemblance 
in the early condition of the mesoderm to the disposition char- 
acteristic of Amphioxus. Traces of evagination of the meso- 
derm along the edge of the notochord, as required by the en- 
terocoele theory, have been described in various Vertebrate 
embryos by a number of authors. These traces are so slight 
and so variable, that I believe they would be considered as 
meaningless variations in the grouping of cells, similar to 
those occurring in other parts of the same embryos, were 
investigators not unduly influenced by the preconceived 
theory. No investigator has claimed, so far as I am aware, 
that any Vertebrate exhibits segmentally arranged mesoder- 
