255 



pole-cells and forming the mesoblast bands, gives origin only to the 

 peritoneal epithelium, the gonads and the longitudinal muscles of the 

 adult body. The other diflerentiated from the ectoderm, not only pro- 

 duces the transitory larval muscles, but also in large part persists in 

 the adult as circular, gut and dissepiment muscles. His earlier work 

 was based on Psygmobranchus and his later on Polygordius and Lo- 

 padorhynchus. In the larval development of all of these annelids 

 he has actually found such a double source of mesoblast and has 

 brought forward a mass of evidence in support of his conclusions. 



On the other hand my observations on Thalassema fall in line 

 with those of a considerable number of students of cell lineage of 

 annelids and molluscs, who have Hkewise found a double source of 

 mesoblast. In addition to the teloblasts derived from the posterior 

 member of the fourth quartet, which give rise to the mesoblast bands 

 (coelomesoblast or entomesoblast), an entirely separate origin has been 

 found for the so-called "larval mesenchyme" from certain cells of either 

 the second or third quartet. The precise origin differs, however, in 

 different cases. Thus, in molluscs, it may arise from one (Unio) or 

 more (Crepidula) cells of the second quartet or from two cells of the 

 third quartet (Physa, Planorbis). In annelids it has been described 

 as originating from cells of the second or third quartet (Aricia) and 

 definitely from three cells of the third quartet (Podarke). There are 

 various other annelids and molluscs whose early development clearly 

 indicates the presence of ectomesoblast. In Dreissensia, Meisenheimer 

 describes several mesoblast cells which, he thinks, must certainly have 

 had an ectodermal origin. The same is the case in Cyclas (Ziegler). 

 The one exception is Capitella, where the "psedomesoblast" is described 

 by Eisig as arising from the posterior cell of the fourth quartet. In 

 Thalassema, however, the ectomesoblast has been found for the first 

 time to arise from early cleavage cells of the first quartet and also 

 from more than one quartet in the same form. 



The fact that in Thalassema the ectomesoblast arises from all the 

 first three quartets and from all the quadrants supports the suggestion 

 made by Wilson i), that in the ancestral mode of development all of 

 the three quartets gave rise to mesoblast as well as to ectoblast . . . 

 "I think, therefore, we need not hereafter be surprised to find the 

 formation of ectomesoblast from more than one of the first three quar- 

 tets, whether in Turbellaria or in higher forms." Conklin^) in dis- 



1) Wilson, E. B., Cell-Lineage and Wood's Hall Ancestral Remi- 

 niscence. Biol. Lectures, 1898, p. 34. 



2) CoNKLiN, E. G., The Embryology of Crepidula. Journ. of Morph., 

 Vol. 13, 1897, p. 151. 



