SCOPE OF PRACTICAL PAINT TESTS 107 



cypress, and Y for yellow pine. In order that a record of each 

 panel might be kept on file, previous to the application of paint 

 to the panels, a complete series of photographs was taken of the 

 panels in sets of four. This work seemed advisable so that the 

 future failure of paint on any one panel, which might be thought 

 due to faulty wood, could be either verified or refuted by a refer- 

 ence to the series of photographs made of the bare panels. 



Construction of Panels. The panels were constructed of 

 Dutch weather boarding, tongued and grooved together in strips 

 of three pieces and capped at the top with a weather strip, 

 forming a finished surface three feet long and fifteen and a half 

 inches high. They were firmly braced together at their backs 



View of Atlantic City Test Fence 



and nailed in such a manner that no portion of the nails would 

 appear on the surface of the panel, thus preventing the staining 

 of the panel from rust. The construction of the framework of 

 the fences at Atlantic City and Pittsburg was of such a nature 

 that they would each accommodate 560 panels of this type. 



Starting of Tests. On account of the lateness of the season, 

 it was found necessary to do the painting of the tests within a 

 building, so that each formula might be subjected to fair and 

 equal conditions of application, thus excluding the blowing of 

 dust or rain upon the painted surfaces, which would have taken 

 place had the panels been painted upon the fence. The painting 

 of the panels began in January, 1908, the temperature within 

 the buildings in which the work was done averaging 50 degrees 

 Fahrenheit throughout the work. 



It was decided to test each formula in three colors, in duplicate, 

 and on each grade of wood, exposing the duplicates on either 



