This research project was conducted in a Health Canada-approved licensed commercial facility in which all cannabis plants were grown using a hydroponic method of cultivation (Punja, 2021c). A range of different genotypes (strains) were grown under greenhouse conditions during the fall–winter months (October–April) and spring–summer months (May–September) over a 3-year period (2019–2022), which generally included four cropping cycles per year (Figure 1). These genotypes represented the most frequently grown at the time of sampling and are described in Table 1 and illustrated in Figure 2. During the winter period, plants were provided with supplementary lighting to achieve approximately 1,600 uEin/m² of photosynthetically active radiation (PAR) using high pressure sodium lamps emitting a wavelength centered at 590 nm. During the summer period, supplementary lighting was provided where needed on overcast days to achieve the required PAR.

All plants were initiated from vegetative cuttings taken from stock (mother) plants of various genotypes grown under greenhouse conditions with a temperature range of 23–28°C and relative humidity of 60–70%. The cuttings were dipped in a rooting powder (containing indoleacetic acid) and inserted into 2.5 cm³ rockwool cubes, which were then placed in trays in a propagation room under high relative humidity (80–90%) and a temperature range of 23–27°C. After 2 weeks, when cuttings had rooted, they were placed into wells cut into 10 cm³ rockwool blocks and transferred to a greenhouse bench for an additional 2 weeks to acclimatize and resume vegetative growth. Following this, they were placed on large cocofibre blocks, one plant per block, and transferred to a large flowering room. After 1 week of growth, the photoperiod was adjusted to 12 h lighting:12 h darkness to induce flowering (Punja, 2021a). The water and nutrient regimes were provided according to commercial growing requirements to ensure optimal growth. The plants were arranged in rows up to 104 m in length containing 216 plants, with 2 rows parallel to each other, and spacing of 0.5 m between plants. Each greenhouse compartment contained up to 27 rows of plants, which were separated by a 1 m wide walkway between rows. The plants were trained and supported by plastic mesh netting that ensured the developing inflorescences remained upright (Figures 1A,B). Pruning of leaves and training of plants were conducted manually according to commercial growing requirements as needed. Plants were grown for 7–8 weeks in the flowering period in each cropping cycle and then harvested. All stems bearing inflorescences were removed manually and the individual flower buds were detached from the stems prior to drying on racks, or the buds were detached after hang drying of the entire stem. Drying was conducted in controlled environment rooms set at 21–23°C and 50–55% relative humidity for 5–7 days.

For analysis of total yeast and mold levels (TYM) in this study, all inflorescence samples were obtained from plants in the final 2 weeks of flower (usually week 7 or 8 in the flowering period) representing different genotypes being grown (fresh samples) or following drying using the rack or hang dry procedure (dried samples). Fresh samples were obtained by randomly removing the terminal inflorescences from plants within selected rows in the greenhouse and placing them inside plastic bags, which were stored overnight at 5°C prior to processing the following morning. Samples were generally collected during the morning hours (from 10:00 am to 1:00 pm). Dried samples were randomly obtained from the drying rooms, with records made of the harvest date and genotype, and stored at room temperature inside plastic bags. All samples were collected in three replicates, each from a different plant for fresh samples, and labeled according to genotype and date of sampling. Samples were collected during the winter and summer growing seasons at various times (usually weekly) during the 3-year sampling period. For data analysis to determine the influence of different variables on TYM levels, the data from replicated samples were assessed using statistical methods as described below.

The tissues used for TYM analysis included fresh and dried inflorescence samples as well as dried cannabis leaves. Samples of 10 gm of fresh tissue or 1 gm of dried tissues were homogenized in a Waring blender to which 100 mL of sterile distilled water was added and the mixture was blended at high speed for 30 s. A volume of 400 mL of water was added, stirred, and an aliquot of 0.5 mL was transferred to agar media and spread using a Q-tip. Three agar media were compared for frequency of recovery of yeasts and molds, colony characteristics, and uniformity across replicate dishes. These media included potato dextrose agar (Sigma-Aldrich Canada, Oakville, ON, Canada) to which 140 mg/L of streptomycin sulfate was added to inhibit bacterial growth (PDA + S), Sabouraud dextrose agar (Fisher Scientific Company, Ottawa, ON, Canada) and tryptic soy agar (Sigma-Aldrich, Oakville, ON, Canada). Duplicate samples of each tissue source were used for grinding and each sample was plated onto four replicate Petri dishes of each of the different media types (n = 8). One batch of dishes was incubated at ambient room temperature (21–23°C) and a duplicate set was placed in an incubator set at 28 ± 0.5°C and both sets were provided with 10–14 h/day of florescent lighting, to replicate flower room conditions. After 5 days, the Petri dishes were assessed for the total number of colonies and the size and morphology of the colonies were recorded. This procedure was repeated for multiple samples of cannabis collected at different times to ensure the consistency of the results. The mean number of colonies in each sample on each of the different media was determined and data are presented with the standard error of the mean (SE, n = 8). Morphological unique colonies were transferred to fresh dishes of PDA + S and following growth over 2 weeks, sub-cultured again to ensure purity, and then subjected to molecular identification as described below.

To compare the accuracy of the grinding and plating assay used in this study with analyses conducted by commercial laboratories that routinely conduct TYM assays, a replicated set of fresh and dried flower samples were subdivided into three sub-samples; two were sent to individual commercial laboratories located in Canada (their identity is kept anonymous). The third replicate was processed using the method described above. TYM counts per g of sample were compared for the three sub-samples. The process was repeated for dried leaf material that was collected from the greenhouse floor.

To identify each morphologically unique colony to genus and species, a PCR method utilizing primers for the ITS1-5.8S-ITS2 region of ribosomal DNA (rDNA) was used (Punja, 2021b). DNA was extracted from mycelium scraped from the surface of colonies on PDA + S using the QIAGEN DNeasy Plant Mini Kit. Aliquots of 1 μL containing 5–20 ng DNA were used for PCR in a 25 μL reaction volume consisting of 2.5 μL 10X buffer (containing 15 mM MgCl₂), 0.5 μL 10 mM dNTP, 0.25 μL Taq DNA Polymerase (QIAGEN), 0.25 μL 10 mM forward and reverse primers, as well as 20.25 μL DNAse- and RNAse-free water (Invitrogen). The universal eukaryotic primers UN-UP18 S42 (5′-CGTAACAAGGTTTCCGTAGGTGAAC-3′) and UN-LO28 S576B (5′-GTTTCTTTTCCTCC GCTTATTAATATG-3′) were used (Punja, 2021b,c). All PCR amplifications were performed in a MyCycler thermocycler (BIORAD) with the following program: 3 min at 94°C; 30 s at 94°C, 30 s at 60°C, 3 min at 72°C (35 cycles); and 7 min at 72°C. PCR products were separated on 1% agarose gels and bands of the expected size (ca. 700 bp) were purified with QIAquick Gel Extraction Kit and sent to Eurofins Genomics (Eurofins MWG Operon LLC 2016, Louisville, KY) for sequencing. The resulting sequences were compared to the corresponding ITS1-ITS2 sequences from the National Centre for Biotechnology Information (NCBI) GenBank database to confirm species identity using only sequence identity values above 99%. A total of 450 fungal and yeast colonies were analyzed. Genbank accession numbers only for the unique fungal species recovered are presented in Supplementary Figure 1.

Air sampling was conducted in the greenhouse environment by placing Petri dishes containing PDA + S close to the inflorescences on plants where they were left with lids removed for 60 min (Figure 1). The sampling was mostly conducted between 11:00 a.m. and 2:00 p.m. The lids were replaced and the dishes were taken to the laboratory for identification of morphologically distinct colonies of fungal species as described above. A total of 500 Petri dishes were used at various times over the duration of this study to characterize the air-borne yeasts and molds. More than 200 individual colonies were selected and analyzed by PCR.

To assess the potential influence of six cannabis genotypes on the presence of TYM in the inflorescences, fresh inflorescence samples were obtained and homogenized and plated at various times as described previously. To determine the influence of time of the year on populations of TYM, replicate samples from the six genotypes were collected from plants growing in June, July, August, and September of 2021 from sequentially harvested crops and plated onto PDA + S. Comparisons were made of the total number of colonies and types of fungi and yeast recovered from combined replicate samples of each genotype (n = 16).

To further determine the possible role of inflorescence leaves on TYM, five genotypes were selected for study. These were “Watermelon Kush”, “Powdered Donuts”, “Death Bubba”, “Pink Kush” and “Jack Herer” (genotype “Black Cherry” was unavailable). Terminal inflorescence stems measuring up to 15 cm in height were removed from the plants and brought back to the laboratory to be analyzed. All leaves emerging from the inflorescences, which included large multi-foliate fan leaves and single unifoliate leaves, were carefully dissected using a scalpel and counted from five replicate samples. To determine the TYM levels in these tissues, 10 g of fan leaves or unifoliate leaves were homogenized and plated onto PDA + S. Colony counts were made after 5 days of incubation (n = 20). The experiment was conducted three times. The mean number of inflorescence leaves in each genotype and the TYM colonies derived from each sample were averaged and expressed with the standard error of the mean (SE, n = 20). To determine if surface-sterilization of inflorescence leaves had an impact on the TYM recovered, dissected leaves from three genotypes were immersed in 0.625% NaOCl for 1 min followed by 70% EtOH for 30 s and then rinsed in sterile distilled water. Samples of 10 g were ground and plated onto PDA + S as previously described. Colony counts were obtained after 5 days of incubation.

To observe the extent to which fungi growing on inflorescences were able to colonize stigmatic surfaces and inflorescence leaves, samples of “Watermelon Kush” and “Powdered Donuts” were harvested from these plants. The latter genotype was visibly colonized by Fusarium sporotrichiodes that grew within the inflorescence tissues from naturally derived inoculum. Tissue segments were prepared for scanning microscopy according to the method described by Punja et al. (2023).

A hand-held temperature and humidity recording psychrometer (Reed 8,706 psychrometer, Reed Instruments, Newmarket, ON, Canada) was used to measure temperature and relative humidity within intact inflorescences on cannabis plants that were approaching harvest. The probe was gently inserted into the inflorescence tissues and held in place for 10 s to obtain a reading. Similar measurements were made of the surrounding environment by exposing the probe to ambient greenhouse conditions adjacent to the row of plants. Temperature and humidity measurements were recorded daily over a 2-week period during June 2021 at four different times: 6:45 a.m., 10:45 a.m., 2:45 p.m., and 6:45 p.m. Cannabis genotype “Pink Kush” was used for these measurements. Additional measurements were made at these four times over a 72-h period for two additional genotypes: “Powdered Donuts” and “Death Bubba” and compared to ambient conditions.

To investigate the influence of circulating air movement on temperature and relative humidity measurements in inflorescences, several fans were mounted above cannabis plants on central posts in the greenhouse and left on for 24 h daily over a 7-day period prior to harvest. Measurements of temperature and relative humidity were made in inflorescences of “Pink Kush” and “Powdered Donuts” under the fans, as well as on control plants not exposed to circulating air in adjacent rows of plants. The trial was conducted during August 2021 and repeated once during January 2022. Inflorescence samples (n = 5) were collected from plants grown with and without fans and TYM levels were assessed by plating as described previously.

Following harvest of cannabis inflorescences, the individual flowers (buds) may be detached from the main stem using a machine that physically strips them off (a process called bucking or destemming) (Caplan et al., 2022), after which the inflorescence leaves are trimmed in a machine that passes the buds through a series of rotating blades (wet trim). The processed buds are dried on a rack on trays placed in a drying room (the wet trim, rack dry method) (Caplan et al., 2022). Alternatively, the entire stem is harvested and hung upside down in the drying room, and following drying, removal of buds from stems and trimming is conducted as for wet trim (hang dry method) (Caplan et al., 2022). To determine TYM levels in inflorescences that were processed by wet trim vs. hang dry, samples (total of 35) were obtained at multiple times from the same greenhouse compartment representing different genotypes to allow a comparison to be made of the resulting impact on TYM. The data from four representative samples are presented.

To estimate the impact of drying on moisture loss and TYM levels, samples of hang-dry inflorescences were taken at daily intervals over a 6-day period from a drying room. Samples were collected from 12 locations within the room and the moisture content was estimated using the method following U.S. Pharmacopeia USP 731.²

Water activity measurements were also made for each sample following the procedure of USP 922.³ TYM levels were assessed in each sample as described previously (n = 12). For each sample, four replicate Petri dishes were used and colonies were counted after 5 days of incubation at 23°C. The procedure was repeated three times using different batches of cannabis samples. The mean number of colonies for each sample was determined and expressed with the standard error of the mean (SE, n = 12).

Inflorescence samples of genotype “Pink Kush” and “Powdered Donuts” processed using the hang dry method were sent to a commercial laboratory for estimation of TYM over a 12-month period from October 2020 to September 2021, representing samples from different cropping cycles. The data from a total of 110 samples were grouped according to the reported TYM levels, which ranged from <100 to >50,000 cfu/g. Because these were individual samples obtained at various time points during the 12-month sampling period, there was no replicated data available for statistical analysis. The individual data were plotted according to frequency of samples present in each TYM category to allow comparisons to be made.

To assess whether there were significant differences among genotypes, inflorescence leaf sterilization methods, impact of air circulation, and drying methods, on TYM levels, statistical analysis was done using RStudio Version 1.3.1093. An analysis of variance (ANOVA) was used to determine significance between the means of TYM from replicated and repeated experiments in these studies. Tukey’s post-hoc test was used to determine which variables were significantly different (p < 0.05) from one another.

The greenhouse environment and development of cannabis plants sampled in this study is shown in Figures 1A,B. The plants were grown in parallel rows and individually spaced about 0.5 m apart. By the time of harvest, the foliage of plants had intermingled, and the inflorescences were fully developed (Figure 1B). Each inflorescence produced multi-foliate fan leaves at the bottom with unifoliate inflorescence leaves emerging at various positions toward the top. A comparison of a fresh cannabis sample with inflorescence leaves and a dried sample in which the inflorescence leaves had been mechanically trimmed is shown in Figure 1C. These tissues were homogenized (Figure 1D) to produce a suspension that was green in color when taken from a fresh sample (Figure 1E) or brown when originating from a dried sample (Figure 1F). The resulting extract (0.5 mL) was plated onto Petri dishes containing PDA + S and spread out using a Q-tip (Figure 1G). A range of fungal colonies could be seen growing after a 5-day incubation period at 23°C (Figure 1H). The air sampling method for detecting yeasts and molds involved placing Petri dishes containing PDA + S on the plant canopy adjacent to the inflorescences at various locations in the greenhouse (Figure 1I). The colonies that emerged were numerous and diverse in morphology (Figures 1J,K) and were counted after 5 days and identified to species level using morphological and molecular methods.

Six genotypes were grown in the same greenhouse compartment and were sampled at the same time to allow a direct comparison to be made under similar growing conditions (Figure 2). These genotypes differed in several phenotypic features that included the overall size of the inflorescences produced (height and width), density of pistils produced (as indicated by the exposed stigmatic surfaces visible on the surface), degree of compactness of the inflorescence (tight vs. loosely structured) and extent of inflorescence leaves formed (unifoliate leaves emerging directly from the inflorescence) (Bernstein et al., 2019). They also differed in their susceptibility to several common diseases (Table 1). There were no differences in the range of THC or CBD levels or presence of the six most common terpenes among the genotypes. However, a correlation was observed between the relative levels of TYM present in inflorescences and the susceptibility of the same genotypes to Botrytis bud rot (Mahmoud et al., 2023) and powdery mildew infection (Scott and Punja, 2022) i.e., higher susceptibility to these diseases was associated with higher TYM levels in the same genotypes under similar growing conditions.

The total number of colonies of yeast and mold present in six fresh and dried cannabis samples plated onto three different media is shown in Figures 3A,B. The standard error bars from four replicates and two duplicates of each sample showed the variation (SE) around the mean. While samples differed in the initial levels of TYM present, the different media showed similar trends. There were consistently lower TYM levels in dried samples compared to fresh samples (Figure 3B). The morphological appearance of colonies on the three-agar media after 5 days of incubation are compared in Figures 3C–E for samples identified as A, B, and E. Colonies on PDA and SDA displayed vibrant colony colors not seen on TSA. On the latter medium, colonies were mostly beige-brown in color and were indistinguishable from each other morphologically. On PDA and SDA, colonies could be identified by colony colors: green (Penicillium), yellow (Aspergillus), black (Aspergillus), and olive-brown (Cladosporium). These identifications were confirmed by PCR analysis (see below). A culture of Fusarium graminearum was plated on all three media for comparison of morphological appearance. On PDA, the colony produced a characteristic red color, which appeared yellow on SDA and white on TSA (Figure 3F). Based on these results, PDA + S was selected as the preferred medium to be included in this study for ease of identification and enumeration of colonies.

To investigate the effect of incubation temperature (23°C vs. 28°C) on TYM growth, duplicate sets of four dishes containing the three media which received extracts from homogenized fresh and dried cannabis tissues were incubated for 5 days and compared for colony numbers and visual appearance of the colonies. Colony size was greater at 28°C vs. 23°C, but overall colony number was generally lower at 28°C (Figures 4A–D). Dried tissues consistently had lower TYM levels compared to fresh tissues, as noted previously. The larger colony sizes which intermingled with each other can be seen on dishes incubated at 28°C (Figure 4E). This made it difficult to discern individual colonies compared to incubation at 23°C, which was used as the preferred incubation temperature throughout the study.

To calculate the levels of TYM present in samples (expressed as cfu/g of tissue), the amount of tissue weighed out (fresh or dried) and the dilution factor was used. Fresh tissue (10 g) or dried tissue (1 g, assumed to contain 10% moisture) were both added to 100 mL of water for blending. Following the addition of 400 mL of water and subsequent plating of 0.5 mL onto Petri dishes, the dilution factor was determined to be 1,000-fold for both fresh and dried samples. Therefore, a single colony represented 1,000 cfu/g while 10 colonies represented 10,000 cfu/g. The range of colonies observed in samples included in this study was <10,000 cfu/g (Figure 4F) to 15,000–100,000 (Figures 4G–K). The colonies were diverse in appearance and following PCR of the ITS1-5.8S-ITS2 region of rDNA, were identified primarily as Penicillium (green or whitish green colonies, Figure 4F), Cladosporium (olive-brown colonies, Figure 4G), Aspergillus spp. (black colonies [Figure 4H], or yellow colonies [Figure 4I], or olive-green colonies [Figure 4J]).

A description of Aspergillus and Penicillium colonies originating from cannabis inflorescences and their spore structures as viewed in the light and scanning electron microscopes are shown in Figure 5. On PDA + S agar, Aspergillus ochraceus, A. niger, and A. flavus produced sufficiently distinct colonies on PDA + S that they could be identified using colony morphology in subsequent tissue plating analyses (Figures 5A–C). Aspergillus ochraceus was the most common species recovered in this study, appearing bright yellow on cannabis inflorescences when placed under high humidity conditions for 48 h (Figure 5D). This species, as well as all other Aspergillus species, produced spores borne in clusters (heads) on erect conidiophores (stalks) that could be seen using the scanning electron microscope (Figures 5E,F). The spore surface was spiny (echinulate) under high magnification (Figures 5G–I). In contrast, Penicillium species growing on cannabis inflorescences appeared whitish-blue (Figure 5J) and the spores were similarly produced in long chains on conidiophores and the spore surface was spiny (Figures 5K,L). In addition to these fungi, a number of yeasts were also recovered from cannabis tissues and are shown in Supplementary Figure 1.

Air sampling Petri dishes were used to determine the effect of leaf litter presence/absence and harvesting activity on TYM levels in the air and on inflorescences. Leaves that were removed during pruning activities were deliberately left on the greenhouse floor to dry over the 7–8 week flowering period in certain locations (Figures 6A,B), and compared to adjacent areas with no leaf litter. Inflorescences were harvested from both areas and analyzed for TYM. When these dried leaf tissues were homogenized and plated (1 g dry weight) to determine the background microbes present in them, developing colonies were identified as A. ochraceus, A. fumigatus, Penicillium, and Fusarium sp. (pink colonies; Figures 6C,D). The impact of harvesting activity, defined as the activity of workers that removed the mesh netting around plants prior to cutting and removing the inflorescence stems, as well as stepping on the dried leaves on the floor where present, on TYM in the air and on flowers, was also assessed. The data show that both the presence of leaf litter as well as the harvesting activity contributed significantly (p < 0.05) to increasing the TYM levels in the air and on flower tissues compared to where no leaf litter was present (Figures 6E,F). The Petri dishes from these studies frequently contained >100,000 cfu/g of tissue and consisted mostly of Penicillium, Aspergillus and Cladosporium species (Figure 6F). There appeared to be a seasonal effect on the TYM levels in these leaf litter experiments. When conducted during May–October compared to November–April (Figure 6G), there were more samples that exceeded the 50,000 cfu/g level in the former. These data were compiled from samples collected at various times and from different genotypes and are presented as a composite within the two sampling periods indicated.

The comparative morphology and visual appearance of the inflorescence samples from six cannabis genotypes included in this study are shown in Figure 7A. When these inflorescence tissues were homogenized and plated during the months of June, July, August and September, the six genotypes were shown to contain significantly (p < 0.05) different numbers of TYM (Figure 7F) as well as diverse fungal morphologies (Figures 7B–E). In all months sampled, the most prevalent fungi were Penicillium spp., A. ochraceus, and Cladosporium spp. based on the morphology of the colonies, which was subsequently confirmed by PCR. In addition, bright pinkish-red colonies of Fusarium sporotrichiodes were seen in July and August (Figures 7C,D). The genotypes “Jack Herer” and “Death Bubba” consistently had the lowest TYM during all months of sampling (p < 0.05) (Figure 7F). The highest levels were seen on “Watermelon Kush” and “Powdered Donuts”.

The general appearance of a large mature inflorescence of cannabis consists of a central stalk around which aggregates of pistillate structures are produced to form a raceme, with larger 3 or 5-foliate fan leaves and smaller unifoliate inflorescence leaves emerging from all sides (Figure 2). To determine the total number of inflorescence leaves produced by different cannabis genotypes, inflorescences of genotypes “Watermelon Kush”, “Powdered Donuts”, “Jack Herer” and “Death Bubba” were obtained from greenhouse-grown plants and brought back to the laboratory. Careful dissection was performed to remove all attached leaves using a scalpel and pair of forceps. An example of a dissected inflorescence of “Powdered Donuts” and dissected leaves laid out is shown in Figure 8A. The inflorescence leaves were enumerated and the results are presented in Figure 8B. They show that there were significant differences (p < 0.05) among the four genotypes in the number of inflorescence leaves, with “Powdered Donuts” and “Watermelon Kush” having more compared to “Jack Herer” and “Death Bubba”. Following this, the fan leaves, inflorescence leaves and remaining inflorescence tissues were homogenized and plated onto PDA + S to determine TYM levels in each tissue type. The results (Figure 8C) show that the majority of TYM were associated with the inflorescence tissues and much fewer numbers were found in the inflorescence leaves and fan leaves. Genotypes “Watermelon Kush” and “Powdered Donuts” had the highest TYM (Figure 8C).

A comparison of the TYM present in inflorescence flower tissues compared to inflorescence leaves of genotypes “Watermelon Kush” and “Powdered Donuts” when plated onto PDA + S is shown in Figures 9A,B. Considerably more colonies were observed on the inflorescence flower tissues compared to the inflorescence leaves in both genotypes. When the inflorescence leaves were surface-sterilized, the TYM levels were significantly reduced in the three genotypes tested but a range of 10,000–20,000 cfu/g could still be recovered (Figure 9C). A comparison of the Petri dishes showing colonies from sterilized and unsterilized tissues from “Watermelon Kush” and “Powdered Donuts” is shown in Figures 9D,E.

The psychrometer probe, when inserted into a cannabis inflorescence for 10 s (Figure 10A), provided readings of temperature and relative humidity that were measured four times a day over a 2 week period and compared to ambient conditions measured within the row of plants (Figure 10B). The humidity within the cannabis inflorescences averaged 75% (+/− 6%) over the duration of the observations, with the highest in the morning (6:45 a.m.) and lowest in the afternoon (2:45 pm) (Figure 10B). By comparison, the daily ambient relative humidity was much lower and the fluctuations were much greater, with an average of 55% (+/− 12%); pronounced dips to as low as 45% were observed at 2:45 p.m. With regard to temperature measurements within the inflorescence, the average was 26°C (+/− 3°C), with the highest in the afternoon (2:45 p.m.) and lowest in the morning (6:45 a.m.). The average ambient temperature over the same period of measurements was 23°C (+/− 3°C); pronounced dips to as low as 19.5°C were observed at 6:45 a.m. (Figure 10B). When the temperature and relative humidity measurements were repeated for two additional genotypes (“Powdered Donuts” and “Death Bubba”) in addition to “Pink Kush” over a 3-day period, the same trends were seen for all genotypes (Figure 10C). The relative humidity and temperature measurements for ambient conditions were always much lower compared to the inflorescences. While there were some differences among the three genotypes, they were not significant when compared to ambient conditions, which were much lower (Figure 10D).

The installation of fans over plants of “Powdered Donuts” and “Pink Kush” (Figure 11A) prior to harvest to provide air circulation for 24 h daily over a one-week period resulted in a significant (p < 0.05) reduction of TYM levels in inflorescences taken from treated compared to control plants (Figure 11B). Measurements of relative humidity and temperature in inflorescences of both sets of plants showed that they were both significantly (p < 0.05) reduced due to enhanced air circulation (Figure 11B). Inflorescences from plants with fans were observed to have a waxier, shinier surface that suggested enhanced cuticular deposition on the inflorescence leaves (Figure 11C); the effect on TYM colonies growing on agar medium can be seen in Figure 11D. Fewer colonies developed from inflorescences exposed to air circulation from fans compared to the controls without fans.

Inflorescences from which individual buds were removed and subjected to a wet trim process to remove inflorescence leaves before being placed on drying racks (Figures 12A,B) were sampled for TYM analysis after 6 days of drying (Figure 12B). For comparison, inflorescences that were not removed and retained on the stem for a hang-dry method (Figures 12C,D), inclusive of a range of genotypes, were also analyzed.

Representative data from four samples are shown in Figure 13A. There was a significant (p < 0.05) reduction in TYM in all samples that were subjected to the hang-dry method compared to the wet-trim method. The appearance of colonies on PDA + S dishes after 5 days of incubation at 23°C is shown in Figures 12E,F. In all four samples, the number of colonies growing on dishes receiving homogenized extracts from wet trim samples was significantly greater compared to those receiving extracts from hang-dry samples. Most of the colonies were identified as Penicillium, Aspergillus and Cladosporium. The effect of the drying process on daily TYM levels in hang-dry inflorescences over a 6-day period is shown in Figure 13B. Following an initial increase in TYM levels during the first 3 days of drying, the populations dropped significantly to around 2,000 cfu/g on day 6. In none of the 35 samples tested was the TYM levels found to be at zero (data not shown). The impact of the drying process on the moisture content of 12 samples taken at daily intervals over 6 days from hang-dry inflorescences is shown in Figure 13C. There was a gradual but significant reduction in moisture content to an average of 12–14% (n = 48) after 6 days.

Inflorescence samples of genotype “Pink Kush” and “Powdered Donuts” processed using the hang dry method were sent to a commercial laboratory for estimation of TYM at various time points over a 12-month period from October 2020 to September 2021. A total of 110 samples were grouped according to the reported TYM levels. The data are presented according to the frequency of samples present in each category. The data in Figure 13D shows the TYM levels ranged from <100 to >50,000 cfu/g. The largest number of samples had TYM levels of 1,000–2,000 cfu/g. Only three samples out of 110 showed a TYM level close to zero, while 5 samples had TYM levels >50,000 cfu/g. When comparing genotypes “Pink Kush” and “Powdered Donuts” for TYM levels over a 12-month period, the latter genotype generally yielded samples with higher TYM levels compared to the former genotype (Figure 13D).