There was no record in any of the data studied of a case where a Type B(U) or Type B(M) package was present on a ship where a fire incident has taken place of sufficient severity to have posed any threat to package integrity. Therefore analyses were carried out to produce an estimate of the frequency of fires on ships of capacity greater than 500 grt of all types which would have been suitable to carry Type B(U) or Type B(M) packages. Thus data for tankers and liquefied gas carriers were not included. However, the usefulness of historical data to the current studies was limited, particularly as a means to estimate fire severity and duration. Such estimates were needed which might be compared against the standard IAEA fire test conditions which apply to Type B(U) and Type B(M) packages. These packages are designed with safety margins in hand relative to the test requirements and may be expected to fail progressively under conditions exceeding those of the regulatory tests.