10 P. R. DAY 



basidiospores . Mutants like those found by Baker and Teo (1966) in P. 

 gvaminis avenae could well be useful. 



Screening for mutants among meiotic products of basidiomycetes in 

 general has received rather little attention. I tried unsuccessfully to 

 recover white or pale basidiospore mutants of Coprinus lagopus by X-ray 

 treatment of immature fruit bodies (Day, unpublished) . 



The recovery of recessive mutants from treated dikaryons raises the 

 question of how they arise. Unless the genetic constitution of the 

 dikaryon is known, induced homozygosity of recessive markers by mitotic 

 crossing over will confound the picture. For virulence markers the ques- 

 tion is complicated by the fact that we know little about the nature of 

 virulence itself. To illustrate my point: avirulence may be the result 

 of a pathogen substance (in this instance a specific inducer) interacting 

 with the host to initiate a response which we will call "resistance". Is 

 virulence merely the lack of such an inducer? If so, then a small dele- 

 tion which removes the gene specifying the inducer, providing it is not 

 lethal, would be expected to produce a virulent phenotype. The recogni- 

 tion and identification of small deletions requires a considerable amount 

 of background knowledge which is not available in the rusts. Even in 

 Venturia inaequalis where some 19 different loci controlling virulence 

 have been identified and, in many cases, mapped (Bagga and Boone, 1968) 

 we cannot say whether virulence is recessive to avirulence at any of these 

 loci, let alone study their interactions, because heterokaryons and 

 diploids of Venturia have not been synthesized. Some answers may come, 

 however, from pseudo-wilds if they are stable for long enough. These are 

 cultures which are disomic for one or more chromosomes usually carrying 

 complementary auxotrophic markers. It would be interesting to compare 

 induced virulent mutants in V. inaequalis with naturally occurring alleles 

 but no such induced mutants are available. 



In experiments where pigment mutants were recovered following treat- 

 ment of rust uredospores with a mutagen (see Table 1) , there is no informa- 

 tion on the genotypes of the treated clones so we cannot distinguish 

 between recombination and mutation. Recessive mutants will be recovered 

 from a homozygous dominant dikaryon if both dominant alleles mutate simul- 

 taneously, if one mutates and the other is lost in a deletion, or if 

 recombination follows mutation of one of the alleles even though they 

 were presumed to be in separate nuclei at the time of treatment. 



NEIOTIC RECOMBINATION 



The sexual stage of a long-cycle rust is obtained by germinating 

 teleutospores and infecting the appropriate host with the resultant basidio- 

 spores. In order to work with single pycnia this must be done in such a 

 way that scattered or isolated pycnia arise, the majority of which are 

 derived from single basidiospores. Flor (1942), in working with basidio- 

 spore infections of M. lini , discarded flax leaves with several pycnia 

 and also allowed 4 to 6 days after pycnia appeared for aecia to develop 

 from undetected multiple infections with compatible basidiospores. 

 Spermatia were transferred from one pycnium to another with a wire loop. 

 In flax rust the aecia form 2 to 5 days later but only 50% of the matings 

 are compatible. Where pycnial infections are long lived or can be clonally 

 reproduced, as in C. vihioola (Patton, 1962), it should be possible to 

 identify the mating type of parental clones, if indeed they are hetero- 

 thallic (Hirt, 1964). The aeciospores are used to inoculate a susceptible 

 host to obtain a uredospore clone which can be tested further. 



