550 MOTOR SYSTEM 



the membrane. The last-mentioned factor is responsible for the 

 unequal distribution of the water of imbibition in different directions 

 within the membrane. This fact, which is of fundamental importance 

 for the theory of hygroscopic movement, was first established by 

 Zimmermann. It is the differences in the power of imbibition in 

 different tangential planes which are of the greatest importance. In 

 the case of a spindle-shaped cell, the most active expansion and con- 

 traction of the walls may take place either in the transverse or in the 

 longitudinal direction. In the former instance it is evident that 

 water of imbibition is most readily introduced (or removed) in the 

 transverse direction, whereas the longitudinal direction is favoured in 

 the other case. 



If we try 'to express this differential imbibition in terms of the 

 molecular structure of the membranes, we are forced to assume 

 that the molecular or micellar groups cohere with different inten- 

 sities in different tangential planes. Where they form continuous 

 longitudinal series water will be incorporated more readily between 

 adjacent longitudinal rows, that is, transversely than between succes- 

 sive micellae of one and the same series that is, in the longitudinal 

 direction. If, on the contrary, the micellae are arranged in transverse 

 series, water is most easily introduced between adjacent transverse 

 rows, that is, longitudinally. For a little reflection will show, that the 

 cohesion cannot be so great between adjacent micellar series as it is 

 between successive micellae of the same fibrillar series, and that, con- 

 sequently, the introduction of water between two series that is, at 

 right angles to their long axes is opposed by a resistance smaller than 

 that which has to be overcome before successive micellae in the same 

 series can be pushed apart. 



The orientation of micellar series in a membrane can be deduced 

 from the arrangement of its slit-like pits or thickening fibres, as well 

 as from any striations that may be visible, since, so far as is known, the 

 two sets of features are always correlated in the same way. If, for 

 example, an elongated cell is found to be furnished with longitudinal 

 or very oblique series of slit-shaped pits, it may be safely concluded 

 that the greatest swelling (and contraction) will take place in the 

 transverse direction. Where, on the other hand, the pits are elongated 

 transversely, or nearly so, the greatest swelling (and contraction) will 

 take place in the longitudinal direction. In either case the maximum 

 extension or contraction may vary, according to Eichholz, between 

 5 per cent, and 20 per cent. Where the histological features give no 

 clue, the axes of maximum expansion and contraction may be de- 

 termined by observing the optical behaviour of the cell- walls in polarised 

 light. 



