The Role of Cell Membranes in Morphogenesis 



157 





pNi 



i 



'A 



Fig. 2. Electron micrograph of tiie mid-bulb region of tiie 

 follicle showing early cell contacts and (probably) contact 

 spread. 



The cells of the matrix contain many mitochondria, 

 many agranular vesicles and large numbers of the 

 dense particles thought to contain ribonucleic acid 

 (3). There are none of the specialised intracellular 

 inclusions which appear later at higher levels in the 

 follicle. The surfaces of the cells are very convoluted 

 and intercellular gaps are common. Numerous small 

 finger shaped pseudopods project from the surfaces 

 and often penetrate deeply into neighbouring cells. 

 Close contact between the cells is limited to a few 

 small areas where the two plasma membranes arc 

 parallel and separated by a distance of about 120- 

 150 A. We conclude therefore that the surface mem- 

 branes of the undifTerentiated cells are flexible and 

 in active motion. Their adhesion is small and contact 

 is both temporary and limited in area. The space 

 between the membranes at these "adhesive contacts" 

 is not by any means empty. A material of rather 

 poor electron scattering power is observed both 

 between the closely opposed surfaces and also spread 

 out over the immediately adjacent surfaces. We sup- 

 pose that this material is the cement responsible 

 for the adhesion of the surfaces. Very little can be 

 deduced concerning its chemical nature. The very 

 slight reaction with the osmium fixative suggests 

 that it is not protein or lipid in nature, but that it 



might be polysaccharide. In view of the important 

 role we shall assign to it in morphogenesis, it is most 

 regrettable that more is not known. 



When the cells further along the follicle are ex- 

 amined the importance of cell contacts becomes 

 apparent (figure 2). The areas of contact spread 

 and. in a /ipper-like fashion, draw the cells together 

 with the result that intercellular gaps are closed and 

 the surface convolutions are smoothed out (figure I ). 

 This development does not take place uniformly 

 throughout the cell population. Contact spread 

 occurs lirsl in the cylinder of cells, which will become 

 the cuticle of the hair, and follows rapidly in those 

 cells, between the cuticle and the outer sheath, which 

 form the three layers of the inner sheath. However, 

 in the cortex, the central cylinder of cells, contact 

 spread does not occur until much later; the cells 

 remain united only locally and intercellular gaps 

 are common. 



Contact spread appears to have important mor- 

 phogenetic consequences, some of which can be 

 deduced from the appearance of the cells of the 

 various layers in the mid and upper bulb. The cells 

 of the cuticle assume an important place in all sub- 

 sequent developments because between them adhe- 

 sive contacts appear early, spread more extensively 

 and seem to be stronger. The cuticle stands out from 

 the surrounding cells partly because of the density 

 of its membranes and also because of the cuboidal 

 shape of the cells in longitudinal section. It appears 

 that the surface adhesion between these cells is great 

 enough to actually modify their shape. By drawing 

 the membranes of contiguous cells together in zipper 

 fashion (4) the mass of previously rounded cells 

 is converted progressively into a columnar layer 

 (figure I ), which divides the advancing stream of cells 

 into two domains whose subsequent developments 

 are in striking contrast. Inside the cuticular "barrier" 

 is the cortex where the characteristic product of cell 

 synthesis is fibrous keratin; outside we tind the inner 

 root sheath where amorphous trichohyaline is the 

 typical product. 



In the cortex contacts are localised and surface 

 activity persists at least as high as the follicular con- 

 striction. The melanocytes, which form the pigment 

 of the hair, are situated largely near the tip of the 

 papilla, and appear able to take advantage of the 

 weak adhesion between the cortical cells to extend 

 their long pigmented processes throughout the cor- 

 tical space. The absence of gaps between the cuticle 

 cells prevents the processes penetrating the cuticle 

 and the inner root sheath beyond. The actual in- 

 corporatiiin of the pigment granules into the 

 cortical cells appears to be a consequence of the 

 continued surface activity of these cells for we can 

 see small pseudopods enveloping the ends of the 

 pigmented processes in a manner suggestive of 

 phagocytosis. Occasionally bundles of granules still 

 enclosed in membranes have been found in cortical 

 cells (I). 



