Microbial response and elimination capacity in biofilters subjected to high toluene loadings

Elimination capacity (EC) is frequently used as a performance and design criterion for vapor-phase biofilters without further verification of the microbial quantity and activity. This study was conducted to investigate how biofilters respond to high pollutant loadings and ultimately how this affects the EC of the biofilter. Two identical laboratory-scale biofilters were maintained at an initial toluene loading rate of 46 g m⁻³ h⁻¹ for a period of 24 days. After the initial biofilm development stage, the loading rates were increased to 91 g m⁻³ h⁻¹ and 137 g m⁻³ h⁻¹, respectively. Following a short period of pseudo-steady state, toluene removal efficiencies rapidly declined in both biofilters, with a concurrent decline in both critical and maximum ECs. The decline was mainly due to deterioration in the biodegradation activity of the biofilm and a decline in the toluene-degrading bacterial population within the biofilm phase. The findings imply that high toluene loadings accelerated the deterioration in overall performance due to a rapid accumulation of inactive biomass. As a result, care must be used when relying on EC values for biofilter design and operational purposes, since the values do not appropriately reflect the temporal changes in biodegradation activity and active biomass quantities that can occur in biofilters subjected to high inlet loadings.     

Effects of UV pretreatment on microbial community structure and metabolic characteristics in a subsequent biofilter treating gaseous chlorobenzene

To provide insight into effects of UV pretreatment on microorganisms in subsequent biofilters, the changes of microbial community structure and metabolic characteristics of biofilters with (UV–BF) and without (BF) UV pretreatment were studied. The respiratory quinone and BIOLOG methods were used to analyze microbial community structure and metabolic characteristics, respectively. The results indicated the quinone profiles, the species of dominant quinone and its molar fraction of the biofilm in both biofilters showed different behaviors. Ubiquinones-8 and menaquinone-9(H₂) was the dominant quinones in BF and UV–BF processes, respectively. The dissimilarity index of two biofilters markedly increased to nearly 60 after turning on the UV lamp. The microbial samples from UV–BF process showed higher metabolic activities of 0.040 cm⁻¹ h⁻¹ than 0.028 cm⁻¹ h⁻¹ in BF process. Moreover, the microorganisms in both biofilters demonstrated distinct metabolic characteristics. Further, the performance of biofilters showed good correlation with microbial community structure and metabolic characteristics.     

The start-up period of styrene degrading biofilters

Styrene vapors from contaminated air were eliminated using long-term adapted mixed microbial culture inoculated on four perlite packed biofilters (serial arrangement, up-flow configuration). During start-up the inlet concentration of styrene rose from 175 to 1300 mg/m³ of total carbon. The total actual residence time in the four biofilters was 24 s. Styrene was successfully degraded by the microbial population in the biofilter. An average of 66% of eliminated styrene was transformed to CO₂. The removal efficiency of the pollutant was, after 18 d of start-up, nearly 85% at an organic load of 170g/m³ per h. The concentration profiles along the bed height were linear for various pollutant inlet concentrations. The total amount of microorganisms in analyzed biomass from the biofilters was about 10⁹ per gram of dry packing mass. The moisture content was around 80% in all biofilters.     

Effects of salinity on microbial tolerance to drying and rewetting

Soil salinity and fluctuations in soil matric potential are stressors for soil microorganisms which, in turn, may affect soil organic matter turnover. In response to salinity and low soil water content, many microorganisms accumulate osmolytes. Therefore, it is conceivable that microorganisms in saline soils are more tolerant to drying and rewetting (DRW) stress than those in non-saline soils. An experiment was carried out with three different salinity levels: electrical conductivity (EC_1:5) 0, 2 and 4 dS m^?1 (EC0, EC2, EC4), and two water treatments: a constantly moist control or two DRW cycles. Respiration as an indicator of microbial activity was measured throughout the 59 days of incubation. At the end of the second dry period (day 35) and at the end of the following moist incubation (day 59), microbial biomass and microbial community structure were determined by phospholipid fatty acid (PLFA) analysis. Increasing salinity decreased microbial activity but did not affect its resistance to DRW. On day 59, cumulative respiration decreased in the order EC0 > EC2 > EC4 with no differences between water treatments. Fungal biomass was negatively affected by salinity at the end of the experiment, while bacterial biomass was unaffected. Microbial community structure in moist treatments differed between salinity levels, with EC4 influencing microbial community structure earlier than EC2. The resistance of microbial communities to DRW stress was salt level dependent; only beyond a critical salinity level adaptation to salt stress was able to reduce the impact of water stress on microbial community structure.     

Elimination of hydrophobic volatile organic compounds in fungal biofilters: reducing start-up time using different carbon sources

Fungal biofilters have been recently studied as an alternative to the bacterial systems for the elimination of hydrophobic volatile organic compounds (VOC). Fungi foster reduced transport limitation of hydrophobic VOCs due to their hydrophobic surface and extended gas exchange area associated to the hyphal growth. Nevertheless, one of their principal drawbacks is their slow growth, which is critical in the start‐up of fungal biofilters. This work compares the use of different carbon sources (glycerol, 1‐hexanol, wheat bran, and n‐hexane) to reduce the start‐up period and sustain high n‐hexane elimination capacities (EC) in biofilters inoculated with Fusarium solani. Four parallel experiments were performed with the different media and the EC, the n‐hexane partition coefficient, the biomass production and the specific consumption rate were evaluated. Biofilters were operated with a residence time of 1.3 min and an inlet n‐hexane load of 325 g m⁻³_reactor h⁻¹. The time to attain maximum EC once gaseous n‐hexane was fed was reduced in the three experiments with alternate substrates, as compared to the 36 days needed with the control where only n‐hexane was added. The shortest adaptation period was 7 days when wheat bran was initially used obtaining a maximum EC of 160 g m⁻³_reactor h⁻¹ and a critical load of 55 g m⁻³_reactor h⁻¹. The results were also consistent with the pressure drop, the amount of biomass produced and its affinity for the gaseous n‐hexane, as represented by its partition coefficient. Biotechnol. Bioeng. 2011; 108:758–765. © 2010 Wiley Periodicals, Inc.     

Shift of microbial community in gas-phase biofilters with different inocula,inlet loads and nitrogen sources

The research on gaseous VOCs biofilters has often concentrated on process optimization. However, the microbial community change upon operating conditions is not well understood. In this study, three lab-scale biofilters treating gaseous toluene were operated for 66 days with different inocula under changes in inlet loads and nitrogen sources. Three biofilters were inoculated with activated sludge, river sediment or microbial consortia, respectively. The microbial community differed a lot initially but gradually deviated toward similar structures with the same dominant microorganisms, i.e. Proteobacteria, Actinobacteria (phylum level) and Rhodococcus,Pseudomonas(genus level). Among three biofilters, the two biofilters inoculated with activated sludge and river sediment showed higher microbial diversity, better VOCs removal performance and higher metabolic activity. Higher relative abundance of Alcanivorax (3% compared with lower than 0.03%), Pimelobacte (0.05% compared with lower than 0.01%)were detected under low inlet load, and Zoogloea(0.1%), Alkaliphilus(0.2%) were detected when the inlet load was increased. the abundance of Pseudomonasdecreased from 14% to 2% when ammonia was used as nitrogen source instead of nitrate, meanwhile the abundance of Bacillus and Gordoniaincreased from 0.01% to 0.05% and 0.8% to 5.8% respectively. Some special organisms were observed i.e. the intestinal microorganism.     

Removal of n-hexane by Fusarium solani with a gas-phase biofilter

A gas-phase biofilter inoculated with the fungus Fusarium solani, isolated from a consortium grown on hexane vapors, was used to degrade this compound. The biofilter, packed with perlite and operated with an empty bed residence time of 60 s, was supplied with hexane concentrations between 0.5 g m⁻³ and 11 g m⁻³. Biofilter performance was evaluated over 100 days of operation. Several strategies for supplying the nutritive mineral medium were assayed to maintain favorable conditions for the fungal growth and activity. The Fusarium system was able to sustain an average elimination capacity of 90 g m⁻³ _reactor h⁻¹ with a maximum of 130 g m⁻³ _reactor h⁻¹ . The mass transfer limitations due to high biomass development in the biofilter were confirmed in batch experiments. Bacterial contamination was observed, but experiments in the biofilter and in batch reactors using selective inhibitors and controlled pH confirmed the predominant role of the fungus. Results indicate that fungal biofilters can be an effective alternative to conventional abatement technologies for treating hydrophobic compounds.

     

A bioactive foam reactor for the removal of volatile organic compounds: system performance and model development

A bioactive foam reactor (BFR), a novel bioreactor operated using surfactant foams and suspended microorganisms for the treatment of gaseous toluene, was investigated to characterize its performance with respect to the mass transfer and biodegradation rates. The BFR system consisted of two reactors in series; a foam column for toluene mass transfer using fine bubbles and a cell reservoir where suspended microorganisms actively biodegraded toluene. In this study, a series of short-term experiments demonstrated that the BFR could achieve stable removal performance and a high elimination capacity (EC) for toluene at 100.3 g/m³/h. A numerical model, combining mass balance equations for the mass transfer and subsequent biodegradation, resulted in reasonable agreement with the experimental findings. At an inlet toluene concentration of 100 ppm_v, the toluene concentration in the liquid phase remained extremely low, indicating that the microbial activity was not hindered in the BFR system. However, the experimental and model prediction results showed that the actual mass of toluene transferred into the liquid phase was not closely balanced with the amount of toluene biodegraded in the BFR used in this study. Consequently, methods, such as increasing the effective volume of the foam column or the mass transfer coefficient, need to be implemented to achieve higher toluene EC and better BFR performance.     

Role of Thiobacillus thioparus in the biodegradation of carbon disulfide in a biofilter packed with a recycled organic pelletized material

This study reports the biodegradation of carbon disulfide (CS₂) in air biofilters packed with a pelletized mixture of composted manure and sawdust. Experiments were carried out in two lab-scale (1.2 L) biofiltration units. Biofilter B was seeded with activated sludge enriched previously on CS₂-degrading biomass under batch conditions, while biofilter A was left as a negative inoculation control. This inoculum was characterized by an acidic pH and sulfate accumulation, and contained Achromobacter xylosoxidans as the main putative CS₂ biodegrading bacterium. Biofilter operation start-up was unsuccessfully attempted under xerophilic conditions and significant CS₂ elimination was only achieved in biofilter A upon the implementation of an intermittent irrigation regime. Sustained removal efficiencies of 90–100 % at an inlet load of up to 12 g CS₂ m^?3 h^?1 were reached. The CS₂ removal in this biofilter was linked to the presence of the chemolithoautotrophic bacterium Thiobacillus thioparus, known among the relatively small number of species with a reported capacity of growing on CS₂ as the sole energy source. DGGE molecular profiles confirmed that this microbe had become dominant in biofilter A while it was not detected in samples from biofilter B. Conventional biofilters packed with inexpensive organic materials are suited for the treatment of low-strength CS₂ polluted gases (IL <12 g CS₂ m^?3 h^?1), provided that the development of the adequate microorganisms is favored, either upon enrichment or by inoculation. The importance of applying culture-independent techniques for microbial community analysis as a diagnostic tool in the biofiltration of recalcitrant compounds has been highlighted.     

Methanol removal efficiency and bacterial diversity of an activated carbon biofilter

Motivated by the need to establish an economical and environmentally friendly methanol control technology for the pulp and paper industry, a bench-scale activated carbon biofiltration system was developed. This system was evaluated for its performance in removing methanol from an artificially contaminated air stream and characterized for its bacterial diversity over time, under varied methanol loading rates, and in different spatial regions of the filter. The biofilter system, composed of a novel packing mixture, provided an excellent support for growth and activity of methanol-degrading bacteria, resulting in approximately 100% methanol removal efficiency for loading rates of 1–17 g/m³ packing/h, when operated both with and without inoculum containing enriched methanol-degrading bacteria. Although bacterial diversity and abundance varied over the length of the biofilter, the populations present rapidly formed a stable community that was maintained over the entire 138-day operation of the system and through variable operating conditions, as observed by PCR–DGGE methods that targeted all bacteria as well as specific methanol-oxidizing microorganisms. Phylogenetic analysis of bands excised and sequenced from DGGE gels indicated that the biofilter system supported a diverse community of methanol-degrading bacteria, with high similarity to species in the genera Methylophilus (β-proteobacteria), Hyphomicrobium and Methylocella (both α-proteobacteria).     

Assessment of microbial populations in methyl ethyl ketone degrading biofilters by denaturing gradient gel electrophoresis

Denaturing gradient gel electrophoresis (DGGE) analysis of polymerase chain reaction-amplified genes coding for 16S rRNA was used to assess differences in bacterial community structure as a function of spatial location along the height of two biofilters used to treat a model waste gas stream containing methyl ethyl ketone (MEK). One of the laboratory-scale biofilters was operated as a conventional continuous-flow biofilter (CFB) and the other was operated as a sequencing batch biofilter (SBB). Both biofilters, inoculated with an identical starting culture and operated over a period lasting more than 300 days, received the same influent MEK concentration and same mass of MEK on a daily basis. The systems differed, however, in terms of the fraction of time during which contaminated air was supplied and the overall operating strategy employed. DGGE analysis indicated that microbial community structures differed as a function of height in each of the biofilters. The DGGE banding patterns also differed between the two biofilters, suggesting that operating strategies imposed on the biofilters imparted a sufficiently large selective pressure to influence microbial community structures. This may explain, in part, the superior performance of the SBB over the CFB during model transient loading conditions, and it may open new possibilities for purposely manipulating the microbial populations in biofilters treating gas-phase contaminants in a manner that leads to more favorable treatment performance.Electronic Supplementary Material Supplementary material is available in the online version of this article at     

Domestic wastewater treatment using multi-electrode continuous flow MFCs with a separator electrode assembly design

Treatment of domestic wastewater using microbial fuel cells (MFCs) will require reactors with multiple electrodes, but this presents unique challenges under continuous flow conditions due to large changes in the chemical oxygen demand (COD) concentration within the reactor. Domestic wastewater treatment was examined using a single-chamber MFC (130 mL) with multiple graphite fiber brush anodes wired together and a single air cathode (cathode specific area of 27 m²/m³). In fed-batch operation, where the COD concentration was spatially uniform in the reactor but changed over time, the maximum current density was 148?±?8 mA/m² (1,000 Ω), the maximum power density was 120 mW/m², and the overall COD removal was >90 %. However, in continuous flow operation (8 h hydraulic retention time, HRT), there was a 57 % change in the COD concentration across the reactor (influent versus effluent) and the current density was only 20?±?13 mA/m². Two approaches were used to increase performance under continuous flow conditions. First, the anodes were separately wired to the cathode, which increased the current density to 55?±?15 mA/m². Second, two MFCs were hydraulically connected in series (each with half the original HRT) to avoid large changes in COD among the anodes in the same reactor. The second approach improved current density to 73?±?13 mA/m². These results show that current generation from wastewaters in MFCs with multiple anodes, under continuous flow conditions, can be improved using multiple reactors in series, as this minimizes changes in COD in each reactor.     

Linking performance to microbiology in biofilters treating dimethyl sulphide in the presence and absence of methanol

The performance and microbiology of two inorganic biofilters treating dimethyl sulphide (DMS) in the presence and absence of methanol was investigated. Addition of methanol was shown to result in an increase in DMS removal for methanol loadings below 90 g MeOH per cubic metre per hour with the optimal methanol loading around 10–15 g MeOH per cubic metre per hour for a DMS loading of 3.4 g DMS per cubic metre per hour, a fivefold increase in the DMS removal rate compared to the biofilter treating DMS alone. Microbial community analysis revealed that the addition of methanol led to a significant increase of up to an order of magnitude in the abundance of Hyphomicrobium spp. in the biofilter co-treating DMS and methanol compared to the biofilter treating DMS alone, whilst there was no significant difference in the abundance of Thiobacillus spp. between the two biofilters. Given the behaviour of the biofilter co-treating DMS and methanol, the magnitude of the increase in Hyphomicrobium spp. in the biofilter co-treating DMS and methanol and the ability of Hyphomicrobium spp. to use both methanol and DMS as growth substrates, it was concluded that Hyphomicrobium spp. were the microorganisms responsible for the bulk of the DMS degradation in the biofilter co-treating DMS and methanol.     

Long-term operation of double chambered microbial fuel cell for bio-electro denitrification

The main aim of this study is to investigate the performance of organic oxidation and denitrification of the system under long-term operation. The MFC reactor was operated in continuous mode for 180 days. Nitrate was successfully demonstrated as terminal electron acceptor, where nitrate was reduced at the cathode using electron provided by acetate oxidation at the anode. The removal efficiencies of chemical oxygen demand (COD) and nitrate were higher in the closed circuit system than in open circuit system. Both COD and nitrate reduction improved with the increase of organic loading and subsequently contributed to higher power output. The maximum nitrate removal efficiency was 88 ± 4 % (influent of 141 ± 14 mg/L). The internal resistant was 50 Ω, which was found to be low for a double chambered MFC. The maximum power density was 669 mW/m³ with current density of 3487 mA/m³.     

Hexane biodegradation in two-liquid phase bioreactors: High-performance operation based on the use of hydrophobic biomass

An innovative operation mode in two-liquid phase bioreactors (TLPB) for the treatment of volatile organic compounds (VOC) was investigated. This mode was based on confining the biocatalytic activity exclusively in the non-aqueous phase (NAP) by using hydrophobic microorganisms. The TLPB was implemented in a 2.5 L stirred tank reactor using 10% (v/v) of silicone oil as NAP and hexane as model VOC. A stable elimination capacity (EC) of 21.0 ± 2.5 g m⁻³ h⁻¹ (corresponding to a removal efficiency of 80%) was recorded for 26 days. The accumulation of inhibitory metabolites resulted in drastic drops in the elimination capacity (EC) and an unstable performance of the system, hexanol being identified as potential inhibitory metabolite. Aqueous culture broth exchange by fresh mineral salt medium at a dilution rate of 0.2 day⁻¹ allowed maintaining a high and sustained VOC removal performance. Dissolved oxygen concentration measurements revealed that the oxidative metabolism was strongly stimulated by the aqueous broth exchange. The temporary blockage of the gas/water/NAP transfer pathway for O₂ highlighted the paramount role of this pathway on the performance of the TLPB based on hydrophobic microorganisms.     

Carbon fluxes, nitrogen cycling, and soil microbial communities in adjacent urban, native and agricultural ecosystems

Urban ecosystems are expanding globally, and assessing the ecological consequences of urbanization is critical to understanding the biology of local and global change related to land use. We measured carbon (C) fluxes, nitrogen (N) cycling, and soil microbial community structure in a replicated (n=3) field experiment comparing urban lawns to corn, wheat–fallow, and unmanaged shortgrass steppe ecosystems in northern Colorado. The urban and corn sites were irrigated and fertilized. Wheat and shortgrass steppe sites were not fertilized or irrigated. Aboveground net primary productivity (ANPP) in urban ecosystems (383±11 C m^?2 yr^?1) was four to five times greater than wheat or shortgrass steppe but significantly less than corn (537±44 C m^?2 yr^?1). Soil respiration (2777±273 g C m^?2 yr^?1) and total belowground C allocation (2602±269 g C m^?2 yr^?1) in urban ecosystems were both 2.5 to five times greater than any other land‐use type. We estimate that for a large (1578 km²) portion of Larimer County, Colorado, urban lawns occupying 6.4% of the land area account for up to 30% of regional ANPP and 24% of regional soil respiration from land‐use types that we sampled. The rate of N cycling from urban lawn mower clippings to the soil surface was comparable with the rate of N export in harvested corn (both ～12–15 g N m^?2 yr^?1). A one‐time measurement of microbial community structure via phospholipid fatty acid analysis suggested that land‐use type had a large impact on microbial biomass and a small impact on the relative abundance of broad taxonomic groups of microorganisms. Our data are consistent with several other studies suggesting that urbanization of arid and semiarid ecosystems leads to enhanced C cycling rates that alter regional C budgets.     

DEWFALL AS A WATER SOURCE FREQUENTLY ACTIVATES THE ENDOLITHIC CYANOBACTERIAL COMMUNITIES IN THE GRANITES OF TAYLOR VALLEY,ANTARCTICA1

Endolithic photosynthetic microorganisms like cyanobacteria and algae are well known from savannas and deserts of the world, the high Arctic, and also Antarctic habitats like the Dry Valleys in the Ross Dependency. These endolithic microbial communities are thought to be at the limits of life with reported ages in the order of thousands of years. Here we report on an extensive chasmoendolithic cyanobacterial community inside granite rocks of Mt. Falconer in the lower Taylor Valley, Dry Valleys. On average, the cyanobacterial community was 4.49 ± 0.95 mm below the rock surface, where it formed a blue‐green layer. The community was composed mainly of the cyanobacterium Chroococcidiopsis sp., with occasional Cyanothece cf. aeruginosa (Nägeli) Komárek and Nostoc sp. Mean biomass was 168 ± 44 g carbon · m^?2, and the mean chl a content was 24.3 ± 34.2 mg · m^?2. In situ chl fluorescence measurements—a relative measure of photosynthetic activity—showed that they were active over long periods each day and also showed activity the next day in the absence of any moisture. Radiocarbon dating gave a relatively young age (175–280 years) for the community. Calculations from microclimate data demonstrated that formation of dew or rime was possible and could frequently activate the cyanobacteria and may explain the younger age of microbial communities at Mt. Falconer compared to older and less active endolithic microorganisms reported earlier from Linnaeus Terrace, a higher altitude region that experiences colder, drier conditions.     

Biofiltration of waste gases containing a mixture of formaldehyde and methanol

Several biofilters and biotrickling filters were used for the treatment of a mixture of formaldehyde and methanol; and their efficiencies were compared. Results obtained with three different inert filter bed materials (lava rock, perlite, activated carbon) suggested that the packing material had only little influence on the performance. The best results were obtained in a biotrickling filter packed with lava rock and fed a nutrient solution that was renewed weekly. A maximum formaldehyde elimination capacity of 180 g m^–3 h^–1 was reached, while the methanol elimination capacity rose occasionally to more than 600 g m^–3 h^–1. Formaldehyde degradation was affected by the inlet methanol concentration. Several combinations of load vs empty bed residence time (EBRTs of 71.9, 46.5, 30.0, 20.7 s) were studied, reaching a formaldehyde elimination capacity of 112 g m^–3 h^–1 with about 80% removal efficiency at the lowest EBRT (20.7 s).     

Performance evaluation and effects of hydraulic retention time and mass loading rate on treatment of woodwaste leachate in surface-flow constructed wetlands

Four surface-flow mesocosm wetlands were operated at different hydraulic retention times during two periods to treat diluted woodwaste leachate that was acidic, of very high oxygen demand, and toxic. Temperature, dissolved oxygen, and redox potential decreased with increasing water depth. However, there was no significant vertical variation in microbial biomass. No significant development in biomass of planktonic microorganisms was found over 6 weeks of initial operation. It took <1–6 weeks for maturation of the biofilm on submerged plant surfaces and the sedimentary microbial community. Mass reduction efficiencies of chemical oxygen demand, and tannin and lignin increased significantly with hydraulic retention time when 10% leachate was fed with tap water. When a more recalcitrant influent was fed, there was a slight increase of reduction efficiency with increasing hydraulic retention time. Reduction rates increased linearly with mass loading rates up to 0.4 kg m⁻³ d⁻¹ chemical oxygen demand and 0.13 kg m⁻³ d⁻¹ tannin and lignin. Precipitation and evapotranspiration had profound impacts on the overall performance and its variability. Mass balance-based operating data of wetlands with a mature microbial community are required for proper performance assessment.     

A comparative analysis of microflora during biofilm development on grape seeds exposed to methanol in a biofilter

The attachment of microorganisms onto biotic surfaces to form biofilm structures on the support media of a biofilter has great impact on biodegradation systems. This study examined the composition of the microbial community that developed on grape seeds (GS) used as support media in methanol degradation biofilters. They were analyzed using conventional microbiology techniques and API galleries. Analysis of microbial counts showed that, in GS before methanol exposure, bacteria and filamentous fungi were predominant over yeasts. In contrast, GS exposed to methanol exhibited more bacteria and yeasts than fungi. Most of the Gram-negative bacteria were the Pseudomonas genus, Bacillus staerothermophilus, Bacillus amyloliquefaciens, and Bacillus pumilus. Rhodotorula mucilaginosa was the primary yeast found. The filamentous fungi Aspergillus sp. Cladosporium cladosporioides, Fusarium sp., and Alternaria sp. were also detected. No Gram-positive bacteria growth was found on GS exposed to methanol. Using scanning electron microscopy, biofilm formation on the GS was examined to reveal the presence of both prokaryotic and eukaryotic microorganisms as biomass accumulation was visible on the seeds. Seeds exposed to methanol for 90 days showed a mature biofilm with cuticle and epidermal layer decline, as well as biofilm dissolution into grape seed integuments.