12 A TEXTBOOK OF THEORETICAL BOTANY 



are characteristic of the pit areas they are not necessarily confined to them, 

 and in some cells with very thick secondary walls, for example in the storage 

 tissues of some seeds, they may be seen to penetrate the whole thickness 



of the wall. 



How these plasmodesma are formed is still uncertain. Formerly it was 

 supposed that they were the remains of protoplasmic fibrils which connect 

 the two daughter nuclei after a division. If this were true they could only 

 occur between sister cells, which is certainly not the case. On the other 

 hand, it is difiicult to believe that such fine threads could bore their way 

 through a wall already formed. It would therefore seem probable that 

 they are established at an early stage, when the cells are separated only by the 

 soft matrix or middle lamella. The presence of protein in the young lamella 

 probably indicates that it is permeated by the cytoplasm. At this stage they 

 may be very numerous, but they occur in groups, and although some may 

 be obliterated as the cell matures, these groups are, in general, the sites at 

 which pits are formed. That the plasmodesma actually provide protoplasmic 

 continuity from cell to cell is shown by the fact that infecting viruses (p. 361) 

 will only pass through cell walls where plasmodesma are present. 



The cellulose wall is doubly refractive, and it may therefore be concluded 

 that it is crystalline in structure. The nature of the molecular arrangement 

 is not, however, indisputably settled. The most generally held view is that 

 which originated in the researches of Nageli in 1863, who called the crystal- 

 line units of the cell wall structure micellae. Analysis of cellulose with 

 X-rays by Preston and others shows a fibrillar structure, due to the linking 

 together of /S-glucose molecules into chains about 500 Angstrom units long 

 by about 50 Angstrom units thick. As the single glucose molecule measures 

 about s Angstrom units in length, a chain of 1,000 would be about 5,000 

 Angstrom units or o-5/x long, which is within the microscopic range. Adjacent 

 chains lie parallel to one another and about 100 of them constitute one 

 micelle. The latter in turn are aggregated into " fibrils," which give rise 

 to the striations visible in the wall, especially after treatment with a swelling 

 agent, the striations being probably slip-planes due to mechanical displace- 

 ment of the fibrils. In addition to cellulose there seems to be some pectic 

 material and some wax present, mixed with the cellulose and probably 

 forming a matrix around the cellulose units. The orientation of the fibrils 

 is spirally around the cell at varying degrees of inclination to the long axis, 

 the thickness of the wall being apparently built up of a series of successive 

 sheets of micelles or of fibre units, the inclination of the spiral being different 

 in successive sheets and changing as the proportions of the cell change during 

 growth. The angle of the spiral is more nearly longitudinal in the thick layer 

 of the wall and more nearly transverse in the thin layers, though it is very 

 variable. The molecular pattern exposed at the inner surface, where it is 

 in contact with the cytoplasm, apparently acts as a molecular mould for the 

 building in of fresh glucose units into the cellulose pattern. 



The growth of the cell in length does not involve any stretching of the 

 existing cellulose chains, for the forces holding their glucose units together 



