Late blight, caused by the oomycete plant pathogen Phytophthora infestans, is a devastating disease of potato and was responsible for epidemics that led to the Irish potato famine in 1845 (refs 1,2,3,4,5). Before the 1980s, worldwide populations of P. infestans were dominated by a single clonal lineage, the US-1 genotype or Ib mitochondrial DNA (mtDNA) haplotype, and sexual reproduction was not documented outside Mexico, the centre of diversity of the pathogen. Here we describe the amplification and sequencing of 100-base-pair fragments of DNA from the internal transcribed spacer region 2 from 28 historic herbarium samples including Irish and British samples collected between 1845 and 1847, confirming the identity of the pathogen. We amplified a variable region of mtDNA that is present in modern Ib haplotypes of P. infestans, but absent in the other known modern haplotypes (Ia, IIa and IIb). Lesions in samples tested were not caused by the Ib haplotype of P. infestans, and so theories that assume that the Ib haplotype is the ancestral strain need to be re-evaluated. Our data emphasize the importance of using historic specimens when making inferences about historic populations.

A wide range of commercially formulated fungicides cause in vitro effects on mating behavior in specific isolates of Phytophthora infestans, the causal agent of late blight of potato and tomato. Four isolates of P. infestans representing each of the four common US genotypes, US-1, US-6, US-7, and US-8 and varying in their sensitivity to metalaxyl, were exposed to a variety of fungicides used to control late blight in petri dish assays at concentrations ranging from 1 to 100 mug a.i./ml. Exposure of each of these normally heterothallic single mating type isolates of P. infestans to 9 of the 11 commercial fungicide formulations tested resulted in the formation of oospores after 2 to 4 weeks. The highest numbers of oospores were formed on media amended with Ridomil 2E (metalaxyl) and Ridomil Gold EC (mefenoxam) at 0.1 to 10 mug a.i./ml, averaging as many as 471 and 450 oospores per petri dish, respectively. Several other fungicides including Maneb, Manzate (Mancozeb), Curzate (cymoxanil + mancozeb), and Acrobat MZ (dimethomorph + mancozeb) also induced oospore formation, producing from 0 to 200 oospores per plate at fungicide concentrations from 0.1 to 10 mug a.i./ml. The metalaxyl resistant isolates formed oospores in response to the fungicides more often than the metalaxyl sensitive isolates. No oospores were formed on media amended with Bravo (chlorothalonil) or Tattoo C (chlorothalonil + propamocarb HCl) and these compounds completely suppressed growth of the isolates at 0.1 and 1 mug a.i./ml. Three metalaxyl resistant A2 isolates mated with both A1 and A2 isolates after exposure to the fungicides Ridomil 2E and Ridomil Gold EC. Alterations in mating type expression were also observed in a metalaxyl sensitive A1 isolate after exposure to Benlate (benomyl). Copious amounts of chemicals are applied annually to potato and tomato production areas to control late blight. Our results indicate that a wide range of chemically diverse fungicides can induce normally heterothallic metalaxyl resistant isolates of P. infestans to form oospores in vitro after short exposures to the fungicides.

Phytophthora infestans causes a destructive disease on tomato and potato. In North Carolina (NC) potatoes are mostly grown in the east, whereas tomatoes are grown in the mountainous areas in the western part of the state. Five genotypes of P. infestans were identified from 93 and 157 isolates collected from tomato and potato over a 5 year period between 1993 and 1998. All isolates collected from potato in eastern NC were the US-8 genotype, whereas only a single isolate was the US-1 genotype. Tuber blight was found on immature daughter tubers in a single field in 1997, however infection on mature tubers was not observed. Within potato fields, a range of sensitivity to metalaxyl was observed among isolates but all were either intermediate or highly resistant to the fungicide. In contrast, isolates from tomatoes included previously reported US-7 and US-8 genotypes and two new genotypes called US-18 and US-19 (A2 mating type, allozyme genotype Gpi 100/100 and Pep 92/100). These genotypes had unique restriction fragment length polymorphism banding patterns, were sensitive to metalaxyl, and have not been reported elsewhere. All genotypes, with the exception of the US-1, were the Ia mitochondrial haplotype. Thus, isolates of P. infestans from tomato were more genetically diverse over time in NC than those from potato and include two new genotypes that are sensitive to metalaxyl.