The concept of the rem first appeared in the literature in 1945,  and was given its first definition in 1947.  The definition was refined in 1950 as "that dose of any ionizing radiation which produces a relevant biological effect equal to that produced by one roentgen of high-voltage x-radiation."  This definition left open the question of what might be considered a "relevant biological effect," and did not prescribe if the roentgen should be measured in air, water, or tissue. Using data available at the time, the rem was variously evaluated as 83, 93, or 95 erg/gram.  Along with the introduction of the rad in 1953, the International Commission on Radiological Protection (ICRP) decided to "continue" the use of the rem. The US National Committee on Radiation Protection and Measurements noted in 1954 that this effectively implied an increase in the magnitude of the rem to match the rad (100 erg/gram).  The ICRP officially adopted the rem as the unit of equivalent dose in 1962 to measure the way different types of radiation distribute energy in tissue, and began recommending values of relative biological effectiveness (RBE) for various types of radiation. [ citation needed ] In practice, the unit of rem was used to denote that an RBE factor had been applied to a number which was originally in units of rad or roentgen. In 1977 the rem was redefined by the ICRP as sievert or J/kg, with the intention that the sievert would come to replace the rem.