b BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 



off from the granular parenchyma in which it, is imbedded, and it appears 

 in most cases to be homogeneous, but in some groups it has a finely 

 granular appearance. These groups of nuclei are quite uniformly dis- 

 tributed in the j^eripheral granular layer over all parts of the body, 

 and similar groups are also found between the muscle fibres of the 

 pharynx and suckers and occasionally in the vesicular parenchyma, 

 especially in the anterior jiortion of the body. 



There are also found in the granular layer of parenchyma somewhat 

 larger oval, pyriform, or spindled-shaped cells (Plate 1, Figs, 4, 7-13). 

 These cells are much less numerous than the subcuticular cells, as a rule 

 lie deeper than they do, and never occur in large groups, though some- 

 times two may lie close together (Fig. 11). The protoplasm stains 

 deeply with carmine and shows large, often flake-like granules. The 

 nuclei are large and clear save for the single nucleolus, which is generally 

 oval in form and unusually large, its length being often nearly as great 

 as the diameter of the nucleus. The long spindle-shaped cells (Fig. 4) 

 are undoubtedly destined to become parenchyme muscle fibres, and the 

 shorter cells, which are usually the smaller ones, may be very well 

 regarded as earlier stages in the formation of such fibres. The long 

 axis of these cells is usually parallel to the surface of the body, as 

 is often the case with the parenchyme muscle fibres. 



A third kind of cell to be found in the granular layer is the giant 

 cell, the so-called " grossen Zellen " of German authors. The cells in 

 Hemiurus which I take to be giant cells resemble somewhat the granular 

 cells just described. Their cytoplasm stains but slightly, and is usually 

 continued into several processes (Plate 4, Figs. 39-41). There is a great 

 variation in the amount of cytoplasm around the nucleus, in some cells 

 the cytoplasmic portion of the cell being very large, while in others it 

 forms such a thin sheath around the nucleus as to be scarcely apparent. 

 According to the recent researches of Bettendorf ('97) these cells are to 

 be regarded as myoblasts, lying, as a rule, at some distance from their 

 differentiated contractile portions, with which they are connected by 

 numerous, delicate, branching processes. If we regard the giant cells as 

 myoblasts of the muscles of the body wall and the granular cells as in 

 process of development into parenchyme muscle fibre, i. e. as myoblasts 

 of parenchyme muscle, it is not at all surprising to find a close resem- 

 blance between them. It is to be noted, however, that the method of 

 formation of the muscle fibres is quite different in the two cases. In one 

 the contractile elements become differentiated in such a manner that the 

 body of the cell lies external to them ; there are usually several contrac- 



