Grafts et al. — 6 — Water in Plants 



BousFiELD and Lowry (1910) from studies on solution volume proposed that liquid 

 water is a ternary mixture of hydrol, dihydrol, and trihydrol. Cooling water results 

 in production of polymerized ice molecules ; heating causes dissociation yielding more 

 steam molecules (hydrol). E^ch change results in increase in volume, this increase 

 being superposed upon the expansion or contraction caused by change in temperature. 

 They showed that the solution volume (change in volume upon addition of a solute) 

 goes through a maximum between 0° and 100° C. when salts having a strong affinity 

 for water are dissolved. 



According to Bousfield and Lowry, this phenomenon depends upon the presence 

 of steam molecules at temperatures above the maximum, and ice molecules at tempera- 

 tures below, both of which are destroyed by the addition of the hydrate-forming solute. 

 The regions of minimum volume at temperatures below and above this point of maxi- 

 mum solution volume represent points of maximum density for the combined water 

 (water of hydration) and a density value of 1.24 for the water of crystallization of the 

 sulfates of certain divalent metals is quoted as an example of the contraction due to 

 intermolecular forces. Such reasoning brings the solution-volume phenomenon into 

 agreement with the theory used to explain the maximum density of water at 4° C. 



Concerning the valence forces of trihydrol, Sutherland (1910) diagrammed to 

 scale the structure of the molecule and calculated from atomic diameters its dimensions. 

 He arrived at a density value of 0.986 whereas that of ice at 0° C. is 0.917. Since the 

 discrepancy is only about 8 per cent, the method suggests that some such grouping 

 may be involved in the formation of ice crystals. Suthejrland maintained that water 

 is a binary system, hydrol, as such, being present in water in quantities too small to 

 detect. Water of crystallization, he suggested, was solid hydrol having a mean density 

 of 1.31. 



In a discussion of ionization, Sutherland (1910) stated "ionization of all electro- 

 lytic solutions at all strengths is complete" and explained the apparent lack of ioniza- 

 tion of so-called weak electrolytes on the basis of mobility of ions. He pointed out the 

 remarkable change in volume occurring when metals of the lithium family combine with 

 halogens, but did not agree with either the theory of partial ionization or the hydration 

 theory to explain the retention of this altered volume by the ions in solution. The 

 mutual energy of the ion and solvent, representing both attractive and repulsive forces, 

 are concerned ; ionic mobility is dependent upon dielectric capacity ; undoubtedly forces 

 between solute and solvent are involved. 



Sutherland attempted an explanation of the unusually high mobilities of hydrogen 



and hydroxyl ions; he calculated the heat of fusion of water of crystallization to be 



1.8 K. cal. per mol ; the heat of vaporization is 5.0 K. cal. This totals 6.8 K. cal. for 



the change from hydrol of vapor to the solid hydrol of crystallization, evidence of a 



profound change in the internal electrical energy of the hydrol brought about by the 



proximity of the electrical fields around the molecules of salt. Nernst (1910) showed 



that the diflferences in the specific heats of steam, water, and ice could be explained by 



the relation 



2H2O = (H20)2 + 2.5 K. cal. (1) 



Professor Walker (1910) summarized the principal conclusions of the symposium 

 with the statement "I should think, as a result of this discussion, one will soon find, 

 even in the textbooks, that whilst ice is trihydrol, and steam monohydrol, liquid water 

 is mostly dihydrol with some trihydrol in it near the freezing point, and a little mono- 

 hydrol near the boiling point." 



Later Work on the Hydrol Theory: — Bousfield (1914) has proposed that the 

 vapor pressure of water has an intimate connection with the proportion of steam mole- 

 cules present, and that the addition of solutes reduces the numbers of both ice and steam 

 molecules in water. This explains the reduction of vapor pressure by solutes. In 

 1917 he attempted to reconcile the osmotic properties of solutions with the structure 

 of water. He attributed osmotic pressure to the thermal agitation of the water vapor 

 molecules, and explained the depression of the vapor pressure and of the freezing point 

 as resulting from a shift in the equilibrium conditions of the liquid water. This con- 

 trasts with the view of van't Hoff, Arrhenius, and many subsequent workers on the 

 osmotic relations of solutions. 



Cryoscopic measurements, used primarily to determine molecular weights, indicated 

 that water in solutions was associated. Studies on specific heats of crystalline hydrates 

 also confirmed this view. Studies on the structure of ice have given association fac- 

 tors of from 3 to 23 while data from vapor density are controversial, some indicating 

 association, others none. A compilation of estimates of water association is given by 

 Barnes and Jahn (1934, Table A). 



