578 CEPHALOCHORDA. 



In the Gapitellidae, a diverticulum forms early at the most posterior 

 part of the archenteron and grows out anteriorly. I believe that this 

 is the rudiment of the accessory intestine, but am not quite certain 

 on this point," With regard to the origin of the gill-clefts, we may 

 assume that the originally blind intestinal diverticula secondarily 

 acquired external openings, so as to allow of the outflow of the 

 respiratory water. The assumption that such perforations occur 

 is supported by the actual presence of pores of communication at the 

 ends of the tentacles in the Actinia, and in the hepatic tubes of some 

 Aeolidae. 



We are not, however, able to regard the Annelidan hypothesis as 

 resting on altogether certain foundations. It seems to us that we 

 have no convincing proof that the many points of agreement which 

 actually exist between the Annelida and the Chordata rest upon true 

 homology. The crucial point of the whole question lies in the 

 decision as to whether it is necessary for us to refer the similar 

 method of segmentation of the body in the two groups to their 

 derivation from a common ancestor. It is evident that, as Bateson 

 (No. 26) has shown, it is not a decisive objection to this assumption 

 that the characteristic segmentation in each of the two groups (the 

 Annelida and the Chordata) has arisen separately, or, in other words, 

 that the common ancestor of the two groups was still uusegmented. 

 Balfour (No. 25) in this connection wrote " that we must look for 

 the ancestors of the Chordata, not in allies of the present Chaetopoda, 

 but in a stock of segmented forms descended from the same un- 

 segmented types as the Chaetopoda, but in which two lateral 

 nerve-cords, like those of Nemertines, coalesced dorsally, instead of 

 ventrally to form a median nerve cord." 



If, after what has just been said, we do not regard the derivation 

 of the Chordata from the Annelida as certainly proved, and institute 

 comparison with the other invertebrate groups, we are confronted first 

 of all by Balanoglossus which, in the possession of gill-clefts in the 

 pharyngeal region and the nerve-strand running dorsally shows 

 striking agreement with the type of the Chordata. This form was 

 formerly thought to be nearly related to the Chordata by Gegenbaur 

 and Huxley and more recently by Bateson, Haeckel, Schim- 

 kewitsch, Morgan, Boule and others. We are far from regarding 

 as established the various homologies assumed by Bateson between 

 Balanoglossus and Amphioxus, and on this point refer the reader to 

 the detailed criticism of the best qualified judge in this matter — 

 Spengel (No. 47, p. 721, etc.) — who denies all relationship between 



