BIVERTICILLATA-SYMMETRICA 643 



ginal areas usually somewhat thinner, plane, heavily sporing throughout, 

 with yellow encrusted mycelium usually evident in marginal and sub- 

 marginal areas; reverse in dull orange or orange-brown shades; penicilli as 

 described above. 



Species description based upon numerous strains examined during the 

 present study, including: NRRL 1048, received in 1938 from J. W. Bowen, 

 Johannesburg, South Africa, as an isolate from coconut matting; NRRL 

 1055, received in 193G from E. F. Sprague, Los Angeles, California; NRRL 

 2124, received in May 1945 from W. Lawrence White, Philadelphia Quar- 

 termaster Depot as an isolate from an army cot in Florida. 



Occasional strains are encountered, of which NRRL 1178 is representa- 

 tive, which differ from the species in producing colonies with little or no 

 yellow encrusted mycelium, and with reverse usually uncolored or slightly 

 greenish. However, central areas of old colonies show abundant yellow 

 encrusted mycelia and rich orange-brown colors in reverse. Conidial struc- 

 tures duplicate those of the species in all essential characteristics. These 

 strains must be regarded as representing a normal type of variation within 

 the species. 



Penicillium variahile, as presented here, includes most of the forms con- 

 tained in Thom's "P. luteum series, non-ascosporic" (1930). As indicated 

 elsewhere, however, this was a very general usage and included strains now 

 assignable to P. rubnim, P. verruculosum, and possibly other species which 

 sometimes produce colonies showing an abundance of yellow aerial my- 

 celium. Furthermore, application of the name P. luteum for non-asco- 

 sporic forms is untenable, unless the origin of such strains can be verified, 

 since this binomial was originally applied to a distinctive ascosporic form 

 (see p. 602). 



We clearly recognize that no sharp line of separation exists between the 

 ascosporic and non-ascosporic series of the Biverticillata-Symmetrica, for 

 undoubtedly strains once ascosporic are commonly isolated as the conidial 

 stage onl3^ Loss of ascospore production often occurs in cultures main- 

 tained in the laboratory, and it is reasonable to suppose that the same oc- 

 curs in nature, although probably less commonly. In a single case the 

 reverse is known to have taken place. NRRL 2125, originally isolated 

 from canvas in Panama, was received in December 1945 from Professor 

 Weston as a non-ascosporic culture typical of the strains assigned by us to 

 Penicillium variahile Sopp. In a subsequent recultivation a few scattered 

 yellow perithecia developed in colony centers on malt agar after three to 

 four weeks. The possibility of contamination was eliminated by careful 

 recultivation. Ascospores were elliptical and conspicuously roughened. 

 The pattern of perithecial initials could not be determined since these de- 

 veloped only in heavily sporulating (conidial) areas. The compact and 



