253 



these cells is carried into the praetrochal region by the shifting of the 

 blastopore and there is, in part at least, transformed into muscles 

 running from the body wall to the oesophagus. The cell in the h qua- 

 drant corresponding to the ectomesoblast cells of the other quadrants 

 takes part, in all probability, in the formation of the storaodceum. The 

 similarity in origin of this part of the larval mesoblast with that of 

 Podarke is, as has already been emphasized, very striking. The anterior 

 cell in both has exactly the same cell lineage (3a 21 2-> 2)1 while the 

 other two difier only in an extra division (3c 2, 1, 2^ 1 ^^^ '^d .^, 2? 2-> i)- 



A second source of functional ectomesoblast is the first quartet, 

 as we might expect from Meyer's observations on the larval develop- 

 ment of other annelids. Shortly after gastrulation seven cells from 

 the praetrochal region sink into the primary body cavity. One of 

 these arises from near the apical plate, another from the b (right, 

 anterior) arm of the cross, and a third probably from the c (right, 

 posterior) arm of the cross. Both the latter, however, are very near 

 the median line. In addition to these there are four lateral cells, two 

 on each side, just above the prototroch and accordingly from the a 

 and c intermediate girdle regions. A large part of the progeny of 

 these cells is converted into larval muscles. The lateral cells remind 

 one, in origin at least, of the neuromuscular Anlagen described by 

 Meyer as furnishing the prototroch with its nerve- and muscle-rings. 

 Such is not their fate, however, in Thalasse ma, for no such rings 

 are present in the trochophore, which in many respects is remarkably 

 simple. Again, the posterior ectomesoblast cells from the third quartet 

 are somewhat similar to the adanal neuro-muscular Anlagen of Poly- 

 gordius. 



The most interesting feature of the cell Hneage is the fact that, 

 in addition to these functional ectomesoblast cells, there 

 are at least ten which are rudimentary and in the end 

 entirely degenerate (Figs. 2 and 3). 



It is not possible within the limits of this paper to do more than 

 indicate the origin of some of these rudimentary cells, as follows. 

 From the first quartet arise at least six, namely, two from the posterior 

 arms of the cross, corresponding in origin with the cells which in 

 Nereis and Amphitrite form glands, and one from each of the 

 intermediate girdle -regions. In Podarke two very similar cells are 

 budded oft" in the posterior arms of the cross, and I am convinced 

 that they suffer a like fate. Such, in fact, is tentatively described to 

 be the case by Treadwell. The rudimentary cells in the intermediate 

 girdle regions are budded off at the second (dexiotropic) division of 



