342 Prof. E. W. Mac Bride, Tlie Artificial Production of 



very nearly realised in the existing genus Gephalodiscus, which belongs to 

 the group of the Hemichorda, and is thus related to the vertebrate phylum. 



Cephalodisms possesses two groups of long ciliated tentacles, and each 

 group spring from a semicircular canal sitixated at one side of the mouth. 

 Gilchrist (8) has shown that the living Cephalodiscus crawls about on its 

 prseoral lobe, and occasionally fixes itself to the substratum by the surface of 

 this lobe, exactly as I have shown that the larva of Astcrina gihhosa does (11). 

 The theory that Asteroidea, and inferentially all other Echinodermata, are 

 descended from an ancestor like Ccplialodiscus, was fully worked out by me in 

 my paper on the development of Asterina gihhosa, and the results of the 

 experiments recorded in this paper confirm my confidence in the soundness 

 of that, theory. 



But, when I described the first two larvje of Echinus with two hydrocoeles 

 which came into my hands (14), I pointed out that the probable descent of 

 Echinodermata from a Cephalodiscus-like ancestor affords at best only 

 a partial explanation of the phenomena with which we are concerned 

 For it is to be noted that spines provided with a nervous and muscular 

 collar are characteristic of the highly developed Echinoidea, and are not 

 found in the more primitive Asteroidea, and the same is true of the highly 

 developed teeth and jaws which constitute Aristotle's lantern. It is certain 

 that the bilaterally symmetrical ancestor possessed none of these structures ; 

 but that they were evolved long after the preponderance of growth of the 

 left hydrocoele over the right one had led to the conversion of the bilateral 

 symmetry into a radial symmetry. The theory that there was a bilateral 

 ancestor of Echinodermata therefore cannot account for the existence of a 

 double set of highly developed spines and dental sacs in the abnormal larva. 



In the paper on these two abnormal larvfe, I suggested that the conversion 

 of a portion of the ectoderm of the left side into highly developed spines, 

 and the development of dental pockets from the wall of the left posterior 

 coelom, were due not to the inborn capacities of these tissues, but to the 

 acquisition of powers conferred on them by some chemical substance, in a 

 word, some hormone, which emanated from the hydrocoele bud. I said, 

 further, that where a similar bud was formed on the right side, it emitted a 

 similar substance, which forced the right ectoderm and the wall of the right 

 posterior coelom to undergo corresponding changes, although they had never, 

 in the history of the race, been modified in this manner. I assumed that the 

 hydrocoele bud was the prime mover in the wonderful array of changes 

 which led to the formation of an Echinus-rudiment, because in normal 

 development it is the first part of this rudiment to appear, but this 

 assumption is proved to be justified by the results of the experiments 



