FEBRUARY 1990                                                     

                                                                  
                   ARSON INVESTIGATIONS                                              
                                                                  
                           By

                    William A. Tobin                                               
                     Special Agent                                                    
                     FBI Laboratory                                                    
                     Washington, DC
                                                    
                                                                  
     Collapsed furniture springs, which have been subjected to 
fire, have been used by arson investigators for many years, both 
as an indicator of arson and/or of a slow, smoldering source, 
such as a cigarette.  However, FBI Laboratory research has 
revealed that collapsed springs are not a valid indicator of the 
presence of an accelerant or of a smoldering source.              

     A review of the literature reveals that there are 
contradictory conclusions regarding the condition of springs 
subjected to fire.  Some sources maintain that collapsed springs 
are, indeed, an indication of a slow, smoldering origin because 
of the consistent heat required to collapse them.  Others argue 
that collapsed springs are positive proof of the presence of an 
accelerant because the temperature required to collapse the 
springs cannot be reached by the burning of normal building 
materials alone.  The contradictory nature of these conclusions, 
therefore, suggests that a detailed metallurgical investigation 
was necessary to determine which theory was valid.                

     When metals are subjected to elevated temperatures, many 
metallurgical parameters and considerations can come into play.  
Metallurgists have long known, for example, that in order for a 
metal to be of equal strength, it must be heated to a certain 
temperature in a certain amount of time.  This can be 
accomplished by applying twice the heat in half the time, or half 
the heat in twice the time.  The FBI Laboratory's investigation 
examined not only the interactions of time and temperature but 
also of chemical composition, fabrication history, and other 
metallurgical conditions.                                         

     Part of the FBI's investigation involved actual test 
burnings of mattresses of varying inner-spring construction.  The 
mattresses were placed in a room full of furniture at the FBI's 
Test Burn Facility in Quantico, VA.  To simulate an 
accelerant-based fire, gasoline was poured on the mattresses and 
ignited.  To simulate a cigarette or other slow, smoldering type 
of fire, 10 ml of propanol was placed on the mattresses and 
ignited.  After the resulting fires were allowed to burn for 
varying amounts of time, they were extinguished and data, 
including photographs, were collected.  The most notable 
conclusion was that within the same mattress, there were both 
collapsed and uncollapsed springs.  Therefore, consistent with 
metallurgical expectations, controlled testing revealed that the 
collapsed, partially collapsed or uncollapsed condition of 
furniture springs is of no value as an indicator of arson or of a 
slow, smoldering fire.