Variation among isolates of the Sirococcus shoot blight pathogen (attributed to Sirococcus conigenus, Sirococcus sp., or one of many synonyms) was studied. Inter-simple-sequence-repeat-anchored polymerase chain reaction fingerprints, nuclear rDNA internal transcribed spacers and 5.8S sequences, and 18S rDNA sequences were analysed. In addition, culture morphology as well as sizes and shapes of conidia produced on seed cone scales in culture were compared. Two main groups were distinguished based on both genetic markers and cultural characteristics. The P group consists of isolates mostly from pine and spruce from Europe and both eastern and western North America; the T group consists of isolates mostly from hemlock from western North America. Results provide a benchmark for comparison of other isolates of these fungi and indicate the need for additional sampling and comparative research of these pathogens.

Interest in development of multicohort stands of red pine (Pinus resinosa) in the Great Lakes region of North America prompted an investigation of the potential impact of the shoot blight pathogen Sirococcus conigenus (syn. S. strobilinus) on understory red pine seedlings. In May 2002 and 2003 healthy, 1-year-old red pine seedlings were planted in the understory of a maturing red pine plantation in northern Wisconsin in an area with a history of presence of this pathogen. Occurrence of shoot blight symptoms was recorded periodically during the summer, and in each year seedlings were harvested in fall and examined for signs of shoot blight pathogens. By fall 2002 and 2003, respectively, shoot blight incidence was 89% and 98% and most seedlings were dying. Pycnidia with conidia of S. conigenus were present on almost all of the symptomatic seedlings. The conifer shoot blight and canker pathogen Diplodia pinea (syn. Sphaeropsis sapinea) was also detected, though less frequently. Pycnidia of S. conigenus tended to be found more frequently on symptomatic current year's shoots than symptomatic previous year's shoots; the opposite was true for pycnidia of D. pinea. Risk from S. conigenus to understory red pine seedlings should be considered in any plans for development of multicohort red pine stands in areas where the pathogen is present.

Abstract. The standard approach for constructing a large-stretch pseudo-random generator given a one-way permutation or given a smallerstretch pseudo-random generator involves repeatedly composing the given primitive with itself. In this paper, we consider whether this approach is necessary, that is, whether there are constructions that do not involve composition. More formally, we consider black-box constructions of pseudorandom generators from pseudo-random generators of smaller stretch or from one-way permutations, where the constructions make only nonadaptive queries to the given object. We consider three classes of such constructions, and for each class, we give a black-box impossibility result that demonstrates a contrast between the stretch that can be achieved by adaptive and non-adaptive black-box constructions.We first consider constructions that make constantly-many nonadaptive queries to a given pseudo-random generator, where the seed length of the construction is at most O(log n) bits longer than the length n of each oracle query. We show that such constructions cannot achieve stretch that is even a single bit greater than the stretch of the given pseudo-random generator.We then consider constructions with arbitrarily long seeds, but where oracle queries are collectively chosen in a manner that depends only on a portion of the seed whose length is at most O(log n) bits longer than the length n of each query. We show that such constructions making constantly-many non-adaptive queries cannot achieve stretch that is ω(log n) bits greater than the stretch of the given pseudo-random generator.Finally, we consider a class of constructions motivated by streaming computation. Specifically, we consider constructions where the computation of each individual output bit depends only on the seed and on the response to a single query to a one-way permutation. We allow the seed to have a public portion that is arbitrarily long but must always be included in the output, and a non-public portion that is at most O(log n) bits longer than the length n of each oracle query. We show that such P.P. Limits on the Stretch of Non-adaptive Constructions 505 constructions whose queries are chosen non-adaptively based only on the non-public portion of the seed cannot achieve linear stretch.

Cedrus deodara is a highly valued conifer widely grown as an ornamental in the Pacific Northwest and southern United States. C. deodara in the Pacific Northwest is normally problem free but occasionally is damaged by dieback of shoot tips, which has been associated with a fungus resembling Sirococcus conigenus. In February 2002, bleeding cankers were observed on 2- to 4-year-old stems of potted nursery stock of C. deodara cv. Karl Fuchs from Clackamas County, OR. Cankers were dark with indistinct margins, shallow, and up to 30 cm long. Infection appeared to have originated with small twigs that had died. Cultures isolated from discolored bark on streptomycin-amended potato dextrose agar (PDA) produced conidiomata with hyaline, fusiform, two-celled conidia typical of S. conigenus (1,3). Inter-simple sequence repeat-polymerase chain reaction fingerprints of an isolate from one of these trees were consistent with the P group of S. conigenus (mostly from hosts in Picea and Pinus spp.) (2). This isolate (02–04, ATCC MYA-2969) was used to inoculate two shoots on each of 12 3-year-old potted deodar cedars in each of two trials. Removing a needle wounded each shoot, and an agar plug colonized with mycelium was placed over the wound and held in place for 2 weeks with Parafilm. Sterile agar plugs were applied to two wounded control shoots on each tree in each trial. After 10 weeks, 25 of 48 inoculated shoots were blighted and drooped with yellow to brown needles that eventually dropped. The pathogen was reisolated from 24 of 25 symptomatic shoots but not from asymptomatic or control shoots. To our knowledge, this is the first confirmed report of S. conigenus as a pathogen of C. deodara. References: (1) P. F. Cannon and D. W. Minter. Taxon 32:572, 1983. (2) D. R. Smith et al. For. Pathol. 33:141, 2003. (3) B. Sutton. The Coelomycetes. Commonw. Mycol. Inst., Kew, Surrey, England, 1980.