4 o8 TRANSFORMATION OF ENERGY 



water must rapidly increase. NAGELI suggested that the force of attraction 

 between the substance and the water decreased relatively more rapidly than 

 that between the micellae, the former being more difficult to separate in inverse 

 proportion to the latter. Thus all the water in the swollen body was not held 

 equally firmly. The particles nearest to the surface of the micella were held 

 most firmly, and as the distance from it increased the mobility of the water 

 also increased, and it was very probable that not all the water imbibed lay within 

 the spheres of attraction of the micellae, but was retained as capillary water in 

 minute spaces arising during the swelling. In fact, REINKE'S (1879) researches 

 on Laminaria have shown very clearly that imbibition water is retained with 

 varying tenacity. In these experiments certainly larger spaces, filled with 

 capillary water, e.g. the cell lumina, play a part. REINKE allowed a por- 

 tion of the blade of a Laminaria, which had absorbed i-026g. of water, to dry 

 in air, and found that it evaporated the following amounts in mg. in successive 

 hours : 148, 115, 105, 91, 74, 84, 68, 57, 51, 51, and, later on, still less. Further, 

 it is possible to express water from a completely swollen portion of Laminaria, 

 containing a large quantity of water, with only slight pressure, but great 

 pressure is necessary to extract water when it is present in small quantity. If 

 a Laminaria consists of 75 per cent, water and 25 per cent, solid, water may be 

 pressed out by a pressure of two atmospheres ; if the proportion of water to 

 solid be 43 per cent, to 57 per cent., a pressure of forty atmospheres is needed to 

 achieve the same result. Again, the swelling may be prevented by great pressure, 

 so that it is very obvious that a great deal of work is accomplished in the pro- 

 cess of swelling. RODEWALD (1895) showed that a pressure of twenty-five to 

 thirty-two atmospheres is needed to prevent dry starch from swelling, and it is 

 known that mechanical operations may be carried out by the swelling of certain 

 bodies, e. g. rocks may be split open by the swelling of wooden wedges, and a 

 skull may be separated into its constituent bones by filling it with peas and 

 allowing them to absorb water. Under these circumstances we may assume 

 that air-dry substances capable of swelling always retain demonstrable quantities 

 of water, and that they are able further to condense water vapour from damp air. 

 Simultaneously with the absorption of water a noticeable alteration takes 

 place in the mechanical characters of the swollen body. If the substance when 

 dry be brittle and only slightly extensible, when swollen it may become pliable 

 and very markedly extensible, and yet it loses its elasticity and its rigidity (under 

 tension and pressure). It is important to note how great are the quantities of 

 water which may be absorbed by a body without entirely losing its rigidity, and 

 without transforming it into a liquid. According to NAGELI the gelatinous cell- 

 walls of certain lower Algae contain only one-half per cent, of dry substance ; 

 but even that is far from being the extreme limit that may be reached, since, 

 according to GERICHTEN (1876), apiin, a glycoside obtained from parsley, begins 

 to lose the capacity for forming jelly only when one part of solid is dissolved 

 in more than 8,000 of water. It is not easy to understand how the characteristic 

 features of a solid are preserved when the individual molecules are separated so 

 far from each other as they are in the cell-walls of the Algae mentioned above, 

 where the molecules must be separated by a distance equal to many times the 

 diameter of the molecule. These considerations force us to accept some other 

 theory of structure than that hitherto held. What we need is a structure which 

 maintains its cohesion sufficiently well even when large quantities of water are 

 imbibed. We should have such a structure if we assume the substance capable 

 of swelling to be permeated by canals so that it consisted of minute particles 

 bound together in all directions, just like the meshes of a net in a plane, or, 

 better still, regard it as the honeycomb of soapsuds, where the walls are formed 

 of a substance capable of swelling while the alveoli are able to take in water. 

 Honeycomb constructions of this kind have been shown by BUTSCHLI (1892- 



