210 PROF. HUXLEY ON THE 



are transitional forms, such as we may expect to find bridging over 

 tlie intervals between all groups, as knowledge widens. In any case, 

 they differ from the foregoing in the development of a segmented 

 mesoblast. In the Ccelenterata, Nematoidea, Turbellaria, Trema- 

 toda, and Eotifera, the mode of origin of the cells which lie be- 

 tween the epiblast and the hypoblast, constitute the mesoblast, 

 and give rise to the connective tissues and muscles of the body- 

 wall and of that of the intestine, is not precisely known. They 

 may take their origin in the epiblast or in the hypoblast, or in 

 both. But, in the Earthworm and Leech, after the epiblast and 

 hypoblast are differentiated, the cells of the latter give rise, by 

 division, to two bands of cells which lie one on each side of the 

 long axis of the ventral face of the worm, and constitute the me- 

 soblast. This becomes marked out by transverse constrictions 

 into segments, and, in each segment, gives rise to all the tissues 

 which lie between the epiblast and hypoblast. The mouth cor- 

 responds with the primitive involution of the Morula ; the anal 

 aperture is a new formation. 



In the Nematoidea and in the lower rhabdoccele Turbellaria, the 

 intestinal canal is a simple tube or sac. But, in some Turbellaria 

 and Trematoda the alimentary canal gives off diverticula, which 

 ramify through the mesoblast and even unite together, giving rise 

 to a gastrovascular canal- system like that of the Ccelenterata. 

 These animals, therefore, have what may be termed an enterocoele, 

 more or less distinct from the proper digestive cavity, but con- 

 nected with it, ramifying through the mesoblast. 



Whether the remarkable group of worms termed Grephyrea by 

 De Quatrefages (and including Sipunculus, Sternaspis, Bonellia 

 &c.) belong to the Archgeostomata, or not, is uncertain, too little 

 being known of the early stages of their development. They ap- 

 pear to me to be closely allied to the Eotifera (compare JBonellia, 

 for example), to the Enteropneusta, and to the Echinodermata ; 

 w^hile Schneider, by his very ingenious comparison of the Phoronis- 

 Wva Actinotrocha with CypJionautes, affords even stronger grounds 

 than those furnished by the structure of Phoronis itself, for sus- 

 pecting that the Grephyrea and the Polyzoa are more intimately 

 connected than has been supposed to be the case. 



It will be observed that the Scolecimorpha present a series of 

 modifications from the unsegmented Turbellaria and Nematoidea, 

 through the imperfectly segmented Eotifera, to the polymerous 

 Oligochseta and Hirudinea, and that the segmentation primarily 

 occurs in the mesoblast. 



