THE ORIGIN AND HISTORY OF ANIMAL LIFE 



605 



Jelly 



Collar cell 



Figure 428. Proterospongia haeckeli, a colonial flagellate that resembles what may have been 

 a link between the protozoans and sponges. 



points in its favor that this theory will be 

 sketched briefly in the following paragraphs 

 as an illustration of the method used in 

 tracing vertebrate descent. 



We have seen that there are a number of 

 subphyla in the phylum Chordata that con- 

 tain animals of a lower grade than the 

 vertebrates. These are: (1) the subphylum 

 Hemichordata (Fig. 203) which includes a 

 few wormlike species; (2) the Urochordata 

 (Fig. 205), which contains a number of 

 saclike animals that exhibit chordate char- 

 acteristic chiefly in the immature stages; 

 and (3) the Cephalochordata, which has 

 but a single genus— Branchiostoma (amphi- 

 oxus) (Fig. 201). 



The amphioxus appears to have much in 

 common with the ancestor of the verte- 

 brates. The essential structural characteris- 

 tics which are possessed in common by the 

 amphioxus and the vertebrates are the pres- 

 ence of (1) a notochord, (2) a dorsal nerv- 

 ous system, (3) a pharynx perforated by 

 gill slits, and (4) a midventral endostyle. 



If we accept the amphioxus as the inverte- 

 brate most closely related to the vertebrates, 

 we may then seek for an ancestor of this 

 form. Such an ancestor is supplied by the 

 tunicates. The adult tunicates (Fig. 205) 

 have retained very few of their primitive 



characteristics, but the larva possesses a typ- 

 ical notochord, a neural tube, a series of gill 

 slits, and an endostyle, which are similar in 

 position and development to these structures 

 in the amphioxus; and it seems probable 

 that the adult tunicate once existed as an 

 animal like the larval tunicate of today and 

 that this remote ancestor was not only the 

 progenitor of the modern tunicates, but was 

 also the direct ancestor of the group to 

 which the amphioxus belongs. 



The search for a vertebrate ancestor more 

 remote than the tunicates leads to a con- 

 sideration of the marine wormlike animals of 

 the subphylum Hemichordata. These species, 

 as previouslv shown (Fig. 203) are pro- 

 vided with clearly defined gill slits, a struc- 

 ture which may be homologous to the noto- 

 chord of the vertebrates, and 4 longitudinal 

 nerve cords of which the dorsal is slightly 

 more pronounced than the ventral and 

 lateral ones. It appears, therefore, that a 

 hemichordatelike animal may possibly have 

 been a vertebrate ancestor of an earlier 

 stage than the tunicates. 



We must look to the larvae of the hemi- 

 chordates for the link which may connect 

 these lowest of the chordates with the other 

 invertebrates, and thus complete our hypo- 

 thetical line of vertebrate descent. The 



