Hyatt.] 128 [March 5, 



the archenteron and the elongated cells are present in the am- 

 pullae. It may be inferred, though direct observation is needed 

 to prove the facts, that the effects of use in catching and assimil- 

 ating food restore the collars and flagella and occasion extra 

 growth and pressure in the bodies of the cells. The theory 

 accounts for the restoration of the elongated forms of the cells 

 and the radial direction of the axes as well as the fact that the 

 imaginary centre instead of the centre or central axis of the body 

 as in the embryo and in the long body of Ascones, is now the 

 axis of the ampulla. The axes of the cells in the gastrula are 

 at first elongated and radial as shown by Schultze in Sycandra, 

 but in this form and in the larvae of Carneospongiae as men- 

 tioned above, they apparently lose their flagella, collars and elon- 

 gated forms in the archenteron, and pass through an epithelial 

 stage ; though possibly such a stage may not be present in some 

 of the Ascones. This change seems at first sight opposed to the 

 theory of pressure, but until we know what conditions surround 

 these cells in the archenteron of the ascula we cannot say whether 

 the advent of epithelial cells is due to heredity or to the cessation 

 of the nutritive function and a consequent relief of pressure dur- 

 ing this stage of growth. The mesenchyme maintains the globu- 

 lar amoeboidal form in all its cells, except in those near the ecto- 

 derm and endoderm. These becoming in part subject to the 

 same pressure as the cells of the ectoderm and endoderm are also 

 in the same degree transformed into membrane cells, and this 

 theory derives additional support from the fact stated farther on, 

 that in Spongilla the inner sides of the dermal cells are flat and 

 the outer sides alone exhibit the capacity to expand and become 

 globular. The general moi'phology, observed movements, and 

 the granular contents show that the cells of the mesenchyme are 

 probably the active agents of growth, and the immediate causes 

 of the pressure, which is apparent in the shapes of the ecto- 

 dermal and endodermal membranes. It still remains to be shown 

 in accordance with the theory that these cells have different 

 modes of division in correlation with the pressure exerted ; but 

 w r ith regard to this point we have no evidence of any value. 

 The researches of Haeckel and Saville Kent have shown that 

 the cells of the endoderm are subject to degeneration wiien taken 

 out of their places in the membrane, and are apt to lose their 



