EVALUATION OF WOOD PRESERVATIVES 125 



hindsight to have something of the character of a judgment before trial; 

 but their bearing on the questions under discussion and their possible 

 effect in retarding the development of more reaUstic methods appear to 

 be important enough to warrant quoting at this time. For example, with 

 special reference to Petri dish agar tests he says: 



'The determination of the toxicity of relatively volatile substances, 

 such as coal tar creosote, is particularly difficult, owing to the control of 

 the loss of preservative during the sterihzation process. In order to pre- 

 vent this loss, it is proposed to place the preservative in small sealed glass 

 ampules, which are later broken to Hberate the preservative to form 

 preservative-agar mixtures of any desired concentration." 



He considers laboratory tests of toxicity of preservatives to have httle 

 or no application in commercial practice, and his opinions are definitely 

 stated as follows : 



' 'Toxicity studies deal only with the poisonous properties of a wood 

 preservative, and therefore they do not give a complete picture of the 

 value of any particular substance as a wood preservative. . . . 



"For commercial work, however, it is of interest to know the amount of 

 material that must he initially injected into the wood to maintain the desired 

 amount of preservative for a definite period of time. (Author's itafics). Lab- 

 oratory studies of the toxicity of wood preservatives do not give this 

 information. Attempts to calculate the amount of material which must 

 be injected into the wood from laboratory studies of toxicity are, there- 

 fore, based upon an erroneous conception of the value of such studies." 



Writing about the laboratory use of impregnated blocks of wood in 

 testing wood preservatives, which was already well under way in Europe, 

 he says that by using wood one may obtain conditions more or less 

 closely resembling but not identical with conditions in actual service; 

 but one would not only have to use a solvent in treating to low retentions, 

 but there would be difficulties in obtaining an even distribution of the 

 preservative in the wood. Furthermore: 



"Getting rid of the solvent would require considerable time, during 

 which a considerable loss of creosote would occur. . . . 



"The composition of the creosote in the impregnated wood after the 

 solvent has evaporated may be quite different from that of the original 

 sample. More important still, the movement of the solvent in the wood 

 during drying would cause an uneven distribution of the creosote." 



The reader will bear in mind that these opinions were expressed in 

 advance of the St. Louis and Berlin conferences on laboratory evaluation 

 methods. Schmitz repeated them essentially in his 1930 paper, saying, 

 for instance: 



"Toximetric values are not in themselves an index of the wood preserv- 



