THE ANCESTRY OF THE VERTEBRATES 525 



are much alike, but from one in which the differentiation of 

 the head somites and their grouping- to form complexes has 

 already progressed quite far, and apparently along the same 

 line. The external resemblance between the heavily armored 

 placoderm fishes and their contemporaries among the arach- 

 noids is certainly striking, and the similarity extends also in 

 the head region to the plates of which the shell or cephalo- 

 thorax is composed (Fig. 143). 



Without subjecting the arachnoid theory to further com- 

 ment other than to say that it has received very little recog- 

 nition, we may pass to that theory which places especial weight 

 upon the notochord, the gill-slits and the dorsal position of the 

 central nervous system, and by means of these has traced the 

 line of vertebrate ancestry through a series of invertebrate 



8 



FIG 144. Amphioxus. 



o, oral hood with cirrhi; x, mouth; g, gonads; y, atriopore; z, anus; t, caudal fin; 

 f, dorsal fin. 



lorms, externally very unlike one another, and each some- 

 what isolated in its systematic position. The first of these 

 in descending series, and representing in a way a simplified 

 vertebrate, stripped of everything save the . essentials, is the 

 now famous Amphioxus (or Branchiostoma) , first de- 

 scribed in 1778 as a shell-less snail, or slug (Limax lance- 

 olatus). It is a shore form, and, with a few specific differ- 

 ences, occurs on almost all coasts. It is one or two inches in 

 length, flattened laterally, and pointed at both ends ; it is thus 

 without a distinct head, but the mouth, which is situated a little 

 ventrally at the anterior end, is equipped with a membranous 

 expansion in the form of a hood. The adult burrows perpen- 

 dicularly into the sand, leaving exposed only the anterior end, 

 and in this temporarily sessile condition it expands the oral 

 hood and collects the debris that drifts past it, much after 



