FORMATION OF THE SKELETON IN THE MADEEPORAEIA. 479 



portions have not morphologically the value of cells, as nuclei 

 can only be shown to be present in some of them. The fol- 

 lowing is M. Krempf's description : 



" In the ectoderm which carpets the skeleton, and which 

 consists of a protoplasmic layer with scattered nuclei, in 

 which it is somewhat difficult to define cellular limits, one 

 sees the skeletal element appear in one place and another 

 under the form of a small mass, readily stainable with nucleus 

 dyes. 



" It presents distinct fibrous structure. Its fibres, clearly 

 individualised, are in general regularly parallel with one 

 another and most often are disposed perpendicularly to the 

 surface of the calicoblastic epithelium. Its shape varies 

 according to the points of the skeleton where one observes 

 it. Sometimes the shape is that of a small parallelepiped, 

 fairly regular and flattened, sometimes it recalls that of a 

 Lepidoptera scale, sometimes even that of a cup compressed 

 parallel with its vertical axis and hollowed by a not very deep 

 cavity." 



The setting of the fibres in the layer external to the ecto- 

 derm and in all the skeletal lamellas was shown in my illus- 

 trations and was very frequently described by me, as also the 

 varieties of form displayed by the skeletal elements at differ- 

 ent parts of the corallum (1. c, pp. 113, 121, 125, 128, 136, 

 137, 138, 144, etc.). To continue M. Krempf's observations : 



'' There is always a small nucleus, homogeneous and highly 

 chromophil, placed at one of the sides of the mass. 



" This element, arrived at its complete development, cal- 

 cifies entirely. It then ceases to form part of the still living 

 cell within which it is developed and adheres to the skeleton 

 with which henceforth it is embodied. But the nucleus of 

 the cell which formed it is not carried away with it. It with- 

 draws on the contrary into the midst of the protoplasmic 

 layer which remains living, and after a period of rest, the 

 duration of which I cannot fix, it presides at the elaboration 

 of a new element similar to the preceding. 



" It is only after having assisted at the formation of a 



