The Arson Set
"The Arson Set," pp 405-411 from
Kirk's Fire Investigation, 4/E by John D. DeHaan, ©1997.
Reproduced by permission of Prentice-Hall, Inc., Upper Saddle River, NJ.
Permission from Prentice-Hall is required for all other uses.
The professional and, to a lesser extent, the amateur arsonist wants
to produce a maximum of destruction no matter how small the first ignition.
To this end, most rational arsonists will be sure there are the fundamental
requirements for a large, spreading fire: ready availability of fuel, proper
ventilation, and reliable ignition. While much attention has been given
to the ignition devices used, the most important considerations to the success
of the fire are sufficient fuel and access of air. Only when delayed ignition
is essential is there any need for unusual methods of igniting the fire.
Many more arsonists choose the time and circumstances of the fire to minimize
their chances of being seen than rely on any unusual (and often undependable)
igniting device. Perhaps the preoccupation of some investigators with such
devices results primarily from their novelty, which adds much interest to
their discovery. In one study by the author, only 12.3% of the arson cases
examined involved the use of any type of ignition device. By far the most
common "set" was the direct pouring of a flammable liquid with
ignition by match, which was found in 61.2% of the cases.27 This trend continued
in cases examined in the author's laboratory during the 1980s. The extent
and nature of preparations, such as building or assembling a device, preparing
the scene (providing kindling of a particular type or arranged in a certain
fashion), providing accelerants, ensuring fire spread, and the like, are
all indications of the organization characteristics of the fire setter.
Such characteristics (organized versus disorganized) can be used to focus
on a possible motive and even develop a personality profile of the arsonist
for use in the investigation.28
Arranging the Fire--Location
Once the target is selected, the arsonist must make several choices that
are critical to the success of the fire. A point of origin must be selected
so that flames will rise into fresh fuel. Although flames will spread fastest
across the underside of a ceiling or overhang, it is usually impractical
to arrange a point of ignition high up on a wall. The floor adjacent to
a flammable wall is the likeliest point for initiating the fire. Walls covered
with gypsum sheetrock do not offer much fuel, and once the paint and surface
paper burn off, the fire will die back rapidly. Unless holes have been broken
or cut through the sheetrock to expose the flammable wall studs or ceiling
joists before the fire, such covering will seriously inhibit the spread
of the fire. Wooden paneling products vary considerably in their fire resistance
but eventually they will all add greatly to the heat release rate of a room.
Furniture may be used if present in the correct type and sufficient quantity.
Older furniture stuffed with horsehair and covered with wool fabrics is
difficult to ignite and does not burn readily once ignited. Nearly all modern
furniture contains large quantities of polyurethane foam padding that can
be readily ignited with a match and that burns with great speed and intensity.
Lighter upholstery fabrics of cotton and synthetics can be flammable, adding
further to the ease of ignition and the fuel load. In a commercial or industrial
building, stocks of finished goods, packing materials, or raw materials
may be moved about to provide a large fuel load in a readily combustible
Within the structure, the point selected for the origin may be limited
by what areas of the building are readily accessible and the "ideal"
spot may be protected by a locked door, gate, or elevator. The arsonist
may then have to select a less perfect location. Basements are very suitable
since the walls and ceilings are often not covered, the entire structure
above the fire providing maximum fuel. A basement location is remote, providing
the best concealment, and the utilities and trash often found in basements
provide many possibilities for an "accidental" fire. Any basement
fire should be investigated with special care, because of all possible locations,
this one is generally most suited to the arsonist's purposes.
Attics, although they provide convenient fuel in exposed structural members,
trash accumulations, and points of possible "accidental" ignition
from exposed wiring are generally not selected as a point of origin. The
availability of fuel is limited to the roof and adjacent structure. Insurance
fraud fires are sometimes started in the attic, however, for that very reason.
The ceiling and roof are destroyed, the building is rendered unusable, and
business is interrupted. When the insurance for these claims is collected,
"repairs" are begun and the building, while being rebuilt, suffers
another more complete fire; the insurance on an "upgraded" building
is then collected.
Closets provide both concealment and some fuel in the form of clothing
stored there. Other than nuisance, spite, or juvenile-set fires, arson fires
in closets are rare. A careful arsonist realizes that the poor flammability
of some clothing, the limited air, and the difficulty of encouraging fire
spread to other rooms make a closet a poor choice for a point of origin.
Clothes closets and beds, when confirmed as points of origin, strongly indicate
a revenge/spite motive, rather than fraud. Although, when smoke damage is
the desired end (for insurance fraud-to gain remodeling expenses, stock
replacement, etc.) rather than a lot of fire, a fire set in a closet may
be the best place to ignite it since ventilation and spread can be controlled.
In an attempt to simulate accidental ignition, some arsonists will kindle
the fire in the immediate vicinity of a fireplace, heater, dryer, or similar
heat-producing appliance. It is hoped that a cursory inspection of the scene
will place the point of origin in the vicinity of such a "normal"
cause, and the fire will be written off as an accidental malfunction of
the appliance. All such fires should be examined carefully, and the investigator
must not prejudge such fires and fall into the arsonist's trap. With very
few exceptions, accidental fires associated with such appliances will start
above floor level and travel upward. A fire set with a flammable liquid
will usually burn down to the floor or even below it. If the diagnostic
signs of fire behavior indicate a very low burn, the entire area should
be examined carefully, as described in Chapter 7.
Arson investigators find themselves investigating a fire set by a truly
professional "torch" only on rare occasions. This is probably
because of two guidelines used by such professionals: (1) the simpler the
better, and (2) use fuels available at the scene. Both guidelines minimize
the materials that must be brought to the scene by the arsonist, and the
fewer the number of times the individual is seen in the area (especially
carrying containers or odd packages) the less the chances of being spotted
by a witness. Simpler sets minimize the chances of elaborate devices failing
to ignite, while they maximize the chances that the entire set will be destroyed
by the fire or even normal suppression activities.
The use of accumulated trash as an ignition point obeys both guidelines.
Trash piles provide an inherent risk of ignition by accidental causes and
many of them are written off as accidental by investigators without careful
examination. In the experience of the author, a small quantity of flammable
liquid on a pile of loose paper and plastic cartons will ensure complete
ignition and yet will be completely undetectable in the postfire debris.
Trash accumulations are frequent in basements, attics, storage rooms, and
around back doors (inside and out). A trash-filled basement is ideal for
producing a large destructive fire. These can be difficult fires to establish
as arson-caused because there is nearly always some possible means of accidental
ignition, such as defective electrical wiring, furnace, water heater, or
appliance. These all must be eliminated as causes in order to prove the
crime of arson. Outside trash piles are often used by the arsonist who does
not have access to the inside of the structure. Since such fire setters
usually do not intentionally provide a path for the fire to spread to the
structure, the fires do not always spread to the interior of the nearby
Many arsonists, especially amateur ones, do not rely on solid fuels available
at the scene and will heavily supplement them with flammable liquid accelerants
to ensure complete ignition. Although such flammables have the ability to
produce a great deal of fire in any given location, there is no automatic
assurance that the flames so rapidly produced will find other fuel in their
path so that they can build into a conflagration. The larger the quantity
of flammable liquid used the larger the initial fire may be, but if a volatile
fuel is involved, the more likely the vapor air mixture will be above the
upper explosive limit (meaning no ignition). A very large flammable liquid
fire may also deplete the oxygen in the room or structure and result in
self-extinguishment before other fuels in the room can be ignited. It is
difficult (and often ineffective) to transport and disperse more fuel than
can be readily carried by hand. Except in rare instances, less than a few
gallons will be used. In instances where large drums of volatile hydrocarbons
are readily available (such as in an industrial situation) or where the
fuel tanks or vehicles can be used as sources, massive quantities of flammable
liquids can be employed. Such unusually high fuel loads can be used to ignite
a secondary source of fire such as a large accumulation of automobile tires
which burn ferociously and are all but impossible to extinguish.
The most frequently used amount of an accelerant is that amount a person
can carry conveniently, without attracting undue attention. A person can
easily carry an amount sufficient for kindling a very large fire, so this
is not necessarily a serious limitation. The amateur nearly always uses
much more of a volatile accelerant than is necessary. By using it judiciously
on top of loosely piled flammables, a person can start a very satisfactory
fire in a building using only a very small quantity of accelerant. If one
were trying to burn a large building in an urban area where detection and
suppression would quickly follow ignition, it may be desirable to employ
a large reservoir of liquid fuel over a period of time. A metal drum of
a suitable fuel such as gasoline, paint thinner, or other petroleum distillate
may be placed at the desired point of origin with enough fuel under and
around it to start the fire and to heat the drum to the point of rupture.
A well-filled drum is not difficult to burst by heating, and the ruptured
drum will then spray fuel into the fire for a long time with virtually no
possibility of extinguishing it by conventional water or fog sprays. There
is always the possibility of an explosion from such a device when flammable
vapors are suddenly mixed into a large fire, but maximum damage is usually
desired by the arsonist. Such a drum, placed at the origin of the fire kindled
with additional fuel may not be suspect as an instrument of arson because
there are many legitimate reasons for the storage of liquid flammables on
most industrial or commercial premises.
The amateur arsonist does not always select the placement of the liquid
accelerant that would provide the most effective ignition. Instead, the
most common situation is the pouring of an accelerant over carpeting, furniture,
and open floor, a pile of paper or trash, or a combination of these. The
pouring of a flammable liquid onto an absorbent surface, such as soil, newspapers,
or carpet, allows some fuel a chance to escape combustion. A stack of newspapers,
for example, soaked in gasoline allows the fire to burn only on the outside
of the stack, with the absorbent papers feeding additional fuel to the fire
in the same manner as the wick of a kerosene lamp. Fuel near the center
of the stack is not exposed to air or even to extreme temperatures and remains
for detection days later, especially if the debris has been water-soaked
to minimize evaporation. A pool of paint thinner in a carpet may burn a
distinct ring with the carpet pile in the center of the ring barely scorched
and a substantial portion of the fuel still present where it was not exposed
to the flame-air interface. Fuel soaked into the soil likewise is poorly
consumed and then only at the surface. Loscalzo, DeForest, and Chao poured
gasoline on plywood, carpet, and soil, ignited it, and tested detectability
of the residue at various time intervals. Depending on the length of time
the gasoline was originally allowed to burn, gasoline residues were identifiable
by gas chromatography as long as 24 hours after burning on carpet and as
long as 162 hours on soil.29
More recent tests showed that small quantities of kerosene (unburned) were
detectable by standard forensic methods on denim jeans only for 24 hours
and for only 72 hours on shoes, stored indoors at 22°C (68°F). Unburned
gasoline was detectable for less than 24 hours on shoes and denim jeans
under the same conditions, but was still detectable on carpet after 7 days
of storage outside at 515°C (4050°F).30
The pouring of a flammable liquid directly on the floor is probably the
most commonly detected form of arson. It is readily detected because a considerable
portion of the fuel remains unburned and what does burn can leave characteristic
burn patterns or diagnostic signs. Poor ignition is due in part to the amateur
arsonist's misconceptions about properties of such fuels. On a smooth, nonabsorbent,
tight floor, little or no damage may result if the burning of the fuel has
not raised the floor covering to its ignition point. (With a direct pour,
the liquid fuel tends to protect the floor covering by evaporation rather
than damaging it.) On a floor with cracks, holes, or joints through which
the fuel can penetrate, holes and lines will burn into the floor in a characteristic
manner. Remember that if the room subsequently goes to flashover, the radiant
heat produced may obliterate many of the more subtle indicators on the floor.
On ground-level floors, sometime enough flammable liquid is poured out that
it seeps completely through the floor and into the soil beneath. In such
cases the wooden subflooring, joists, and soil should be checked for the
presence of volatile fuels. Another common mistake for the amateur is to
dump the fuel in the center of the room without ensuring that the flames
will reach furniture, walls, or other fuels to keep the fire going.
Multiple ignition points are common in arson fires. Not satisfied to
gamble on a single point of ignition, the arsonist thinks that "more
is better," with the intent of producing a larger fire that spreads
more quickly and widely than that from one point. These multiple points
are usually connected by a trailer of flammable liquid, smokeless gunpowder,
rags, twisted ropes of newspaper, waxed paper, or wood packing material
(excelsior). In a few cases, a trailer of flammable liquid has been poured
from electrical outlet to outlet along a baseboard, apparently in hopes
of simulating a massive electrical fire. Although multiple sets can provide
a more complete destruction of the target, they present many problems that
the arsonist would just as soon avoid. They require additional materials
and fuels; there is also the chance that one of the elements will fail,
leaving intact sets to be discovered. It requires more time to distribute
fuel in several sites than one and, while the distribution is proceeding,
vapors from the earlier sets may be forming an explosive mixture that may
deflagrate, involving the arsonist. The shoes and clothing often are splattered
with the fuel and may ignite. Especially if the fuel has a high volatility,
the hazards from these sources are likely to outweigh any advantage that
is gained from multiple points of ignition. The time required also prolongs
the chances of the earlier sets being detected and the perpetrator being
spotted while still at work. A low-volatility fuel such as kerosene may
be suitable for such operations, but ones like gasoline or lacquer thinner
can be extremely dangerous.
It should be noted that multiple points of ignition are not necessarily
proof of arson. Heat from a general fire can melt and ignite the polystyrene
light diffusers used in ceiling lighting fixtures or the asphalt from a
collapsing roof. These flaming "liquids" can produce multiple,
apparently unconnected, points of origin throughout a large area. Such multiple
points of origin have been mistaken in the past for arson sets.
Of all the remote-ignition devices known, the Molotov cocktail is the
most common. It consists of a breakable glass container of a flammable liquid
which depends on a lighted wick or a chemical incendiary device to provide
ignition. Its effects are maximized when it can be thrown against a hard
surface (wood is often not hard enough) to break the bottle and mechanically
atomize the gasoline into an explosive mist. The ignition then produces
a fireball that can spread at very high speeds (up to 10 meters per second).
Although a Molotov is usually ignited by its own flaming wick, hypergolic
mixtures have been used to avoid the use of an open flame. A packet containing
sugar and potassium chlorate is taped or tied to the outside of the bottle,
while sulfuric acid is added to the gasoline inside. Upon breaking, the
acid reacts with the sugar and chlorate to initiate a very hot flame which,
in turn, ignites the gasoline. Although the production of a thickened "napalm"
has been recommended in terrorist literature, the minimally enhanced burning
effects are generally not worth the extra time and effort required to mix
the gasoline with soap flakes or paraffin wax.
Although the use of such a device is convenient from the standpoint of
the arsonist, its easy detection at a fire scene makes it less than ideal.
A broken bottle, often a broken window, and perhaps eyewitnesses accompany
its use. Tracing it to the arsonist may be difficult but often it is not.
The broken bottle will usually be at the bottom of the fire where it is
exposed to the minimum amount of heat, so fire damage to the fragments is
minimal. The portions of the bottle bearing the most information about its
origin-the neck with its labeling and the base with its cast-in production
data (date and place of production and product code)-are the pieces most
resistant to mechanical and fire damage and usually survive intact. The
investigator should endeavor to collect all or nearly all of the pieces
to allow reconstruction of the device for identification or even a search
for fingerprints. (As noted previously, the oily components of latent prints
are somewhat soluble in gasoline and may be washed off by the accelerant
as it seeps from a broken container, but new techniques now promise recovery
of prints even in the presence of gasoline.)
Plastic bottles or containers filled with fuel have been encountered
in a number of fires. These may be placed over room heaters to soften and
melt, but are more commonly surrounded by loose paper and trash that is
set afire to provide a means of softening them and igniting the spilled
contents. If properly supported by combustibles, these containers can burn
sufficiently to make identification impossible.
One variation on this is the use of a toy balloon or plastic bag containing
gasoline that is suspended by a string and set swinging over a candle. Once
the oscillations die out, the balloon stops over the candle, which then
melts a hole in the bag; the fuel released quickly spreads the fire to the
surrounding trash. Although effective if small quantities of accelerant
are used, the amateur will often use a large trash bag with lots of fuel
that is more likely to extinguish the candle as it rushes out than be ignited
Type of Accelerant
As we have seen previously, the type of flammable liquid used as an accelerant
is determined in great part by the experience, motivation, and intent of
the arsonist. In a premeditated structure fire set by someone for money
(note that this does not imply the person is a professional torch), some
thought will be given to selecting an appropriate fuel. Gasoline is by far
the most commonly detected accelerant because it is an efficient and easily
ignited flammable liquid and because it can be purchased and transported
even in large quantities without arousing the suspicions of the seller or
authorities. Paint thinners and kerosenes are the next most common fuels.31 They are effective fuels
and can be purchased in moderate quantities without suspicion because there
are legitimate uses even in the private home. Turpentines and lacquer thinners
are almost never encountered as premeditated fuels. They are very costly
and not often used in bulk by private individuals. The chemistries of new
paints make their use more of an exception than the rule it once was. Most
of the paints sold today for household use are the water-based latex type
which do not use ignitable liquids as thinning or cleaning agents.
Although most arsonists seem to give little thought to the accelerant's
volatility, it is a property that has great bearing on the fuel's suitability
to the task. Extremely volatile fuels such as gasoline, lacquer thinner,
or alcohols may evaporate excessively at high temperatures, interfering
with the intended ignition and spread of the fire. The vapors generated
can travel considerable distances and produce explosive mixtures in the
vicinity of distant "accidental" sources of ignition, most typically
the pilot light on the water heater or furnace. This makes it necessary
to set the accelerant very quickly and to ignite it from a safe distance,
thereby reducing the arsonist's control of the set. Kerosenes, paint thinners,
and proprietary lantern fuels are easier to use because of their lower volatility.
The varieties of lantern and campstove fuels can vary considerably in their
volatilities and flash points. Because they may contain heavier hydrocarbons
with commensurately higher boiling points, however, they are more resistant
to evaporation and will remain in the debris longer as residues than will,
say, lacquer thinners.
At the opposite end of the petroleum distillate scale is fuel oil. It
is very difficult to burn except when it is dispersed into a mist and well
mixed with air. When applied to paper or trash, it can burn vigorously as
a thin film and thereby add to the fuel load, but it will not be readily
ignited as a pool.
Alcohols, both methyl (wood) and ethyl (grain), are as volatile as gasoline
and carry the same hazards. Because of their flammability and their reduced
chances of detection because of their miscibility with water, they would
appear to be good choices as flammable accelerants. They are not, however,
efficient fuels, and it takes a larger quantity of burning alcohol to generate
sufficient heat to bring other nearby combustibles to their ignition points.
They will still burn when mixed with some water, but with a lower risk as
well as effectiveness. Potable liquors of 90 to 100 proof (45% to 50% alcohol)
can be used as accelerants, though not very effective ones.
The foregoing discussions deal with premeditated fire sets involving
accelerants. Depending on the motivation and opportunity of the arsonist,
such fuels may not be acquired beforehand. When it comes to dealing with
fires set using materials already at the scene, ingenuity knows no bounds.
Glues, dye solvents, rubber cements and their solvents, mop cleaners, uncured
polyester resin, paints, brush cleaners, and the like, are common in homes,
schools, and businesses and all have been used as effective arson accelerants.
It is under these circumstances that an inventory of the scene and a thorough
interview with owners, tenants, and maintenance people may be vital to the
success of the inquiry. Although it is commonly known that the duplicating
fluid used in the old-style "spirit" duplicators or mimeograph
machines is highly flammable methyl alcohol, it is not commonly known that
the copier toner used in some modern office "dry" copiers contains
a petroleum distillate or isoparaffinic blend, similar to paint thinners.
Many offices have cases of this fluid stored where it is easily accessible
to burglars intent on erasing signs of their presence. The same holds true
for commercial insecticides that are often dissolved in a light paint-thinner-like
petroleum distillate. When such unusual fuels are suspected by the investigator,
these suspicions should be relayed to the laboratory examiner along with
control samples of all possible fluids.
Kirk's Fire Investigation, 4/E by John D. DeHaan, ©1997.
Reproduced by permission of Prentice-Hall, Inc., Upper Saddle River, NJ.
Permission from Prentice-Hall is required for all