wood and between the polysaccharides and water, and the 
necessity of rupturing these bonds when passing from one 
condition to the other (Kretovich, 1966). Although hydrogen 
bonds are rather weak (ca. 105 kcal./mole) in comparison to 
covalent or coordinate bonds, the cumulative effect of rela- 
tively large numbers of them does impart considerable adhe- 
sive force. This idea may be extended further to account for the 
initial fixation of soluble silica to the woody template. Silicic 
acid, by virtue of its four hydroxyl groups per molecule — or 
even more in the case of polysilicic acid — might be expected to 
bond securely to the organic matrix because of its potential for 
forming multiple hydrogen bonds. 
There are a number of observations which appear to be in 
accord with the theory that bonding, probably hydrogen bond- 
ing, 1s involved in the silicification of wood. 
First, coalified wood does not silicify as well as relatively 
unaltered wood. This might be explained by the appreciable 
loss of active sites, i.e. functional groups capable of forming 
bonds with silicic acid, from the former during coalification. In 
fact, Heathcoat and Wheeler, who determined the hydroxyl 
percentage in coal as function of rank, failed to identify hyd- 
roxyl oxygen beyond the rank of 83% carbon (Van Krevelen, 
1961). 
Secondly, the quality of replication of woody structure in 
petrifactions mineralized with carbonate and sulfide are quite 
inferior to those formed from silica. Also, they are of less 
common occurrence. In this case, the explanation may lie in 
the inability of the carbonate or sulfide ions to establish hydro- 
gen bonds with the ligno-holocellulosic components of wood 
and, therefore, deposition is chemically independent and non- 
oriented. 
Thirdly, silicified woods often occur in a matrix of encasing 
sediment which is not consolidated with silica cement, suggest- 
ing Once again that the differential affinity might be explained 
by the potential capacity for wood to bond with and therefore 
serve as a ‘sink’ for silica. 
As a final note, it should be mentioned that ina recent study 
of silica in living plants, Scurfield et al. (1974) observed that 
silica particles often occur with starch grains or with 
polyphenolic material in the same cell, and suggested a casual 
relationship to be operative. Perhaps here too the association 
and mechanism of accumulation might be explained through 
hydrogen bond formation. 
Pie, 
