predominant strain. It is recognized that another strain of the same virus with an 
altered biohazard potential may justifiably be assigned to a different level." 
(SALS, 1980*) 
Had SALS been aware of results suggesting that the SFV strain involved in the 
Tubingen incident was one of increased virulence, it might well have chosen to 
classify the prototype strain of SFV as a Level 2 agent, while placing SFV strains 
of enhanced virulence, such as the multiply passaged Tubingen strain, at Level 
3. 
The correspondence of the SALS group and interviews with former SALS 
members both indicate that the Tubingen fatality was the principal concern of the 
committee regarding the proper classification of SFV. SALS recognized this case 
as "...a unique incident in a long history of work with SFV under minimal 
biocontainment conditions. "(SALS, 1980*) But in the absence of information 
clarifying the circumstances of this anomaly, the committee felt compelled to 
classify SFV at BL3. 
The 1980 SALS recommendations on the containment and practices for 
arboviruses were subsequently adopted, largely intact, into the CDC/NIH 
biosafety regulations in effect today. But the ambivalence of the Subcommittee 
in 1979 still echoes in its recommendations of 1993: "SALS continues to classify 
SFV as a BL3 virus, with the caveat that most activities with this virus can be 
safely conducted at Biosafety Level 2."(BMBL, 1993) 
4. The SALS Survey on Laboratory Experience with SFV 
SALS conducted, between 1976 and 1978, two worldwide surveys of laboratory 
experience and practices with approximately 400 arboviruses (arthropod-borne 
viruses). As described in the report summarizing the results of these surveys 
(SALS, 1980*): 
A checklist questionnaire was developed by SALS to evaluate 
existing and past levels of biological containment in arbovirus 
research facilities, the frequency with which catalogued viruses 
have been handled at each laboratory, the numbers of people at 
risk, the numbers of overt and subclinical infections, and the 
probable routes of infection. 
The first survey was mailed, in 1976, to 494 laboratories worldwide. 158 
laboratories responded, 48 of which employed SFV. A second survey, in 1978, 
was mailed to 585 laboratories to seek additional information. No response was 
requested unless new information on virus use was available. Ninety-four 
laboratories responded, of which 16 had used SFV. 
Recombinant DNA Research, Volume 18 
[465] 
