16 MORPHOGENESIS IN CILIATES 



small field of granules, which disappear. The row a gives 

 rise to another ellipsoidal field. 



While these processes are on the way, the middle part of 

 the rows 8 and 9, kinetodesma and kinetosomes, disappears. 

 The kinetosomes of the anterior part undergo one division. 

 But the tomite is not yet ripe. In order to be ripe, the two 

 granules of the row 9 have to divide once more, thus giving 

 four rows of granules. 



During the early period of tomitogenesis cilia have ap- 

 peared. They are not represented on the drawings. They 

 are ordinary cilia, all over the body. However, the rows 8 

 and 9 and the field a produce specialized thigmotactic cilia 

 which have the property of ceasing their movements when 

 coming into contact with a solid surface, thus assuring ad- 

 herence to the substrate. This important patch of special- 

 ized cilia will allow the tomite to attach itself and to encyst 

 on the integument of the crab. 



Let us now imagine a morphologist looking at the living 

 tomite and considering these specialized cilia without know- 

 ing their origin. His conclusion would be that the forma- 

 tion of this specialized structure is the result of the action 

 of an organizer. 



Dalcq and Pasteels, explaining the neural induction, pos- 

 tulate a multiplication of particles in the neural system 

 under the influence of an organizer, and compare these gran- 

 ules with a virus. An agent possessing the ability to induce 

 multiplication of such granules would be, according to 

 Dalcq and to Brachet, an ''evocator." This conception ap- 

 plies very well to the formation of the thigmotactic field, 

 with the difference that we see the granules really multi- 

 plying. There is of course no reason to refer to self -repro- 

 ducing cytoplasmic units as viruses. 



Before escaping the cyst, the tomite has to form tricho- 

 cysts, but this phenomenon will be examined later on. 



When analyzing development, embryologists have found 

 that the original control of differentiation in all cases ap- 



