What Is Environmental Monitoring in Biomanufacturing?
Environmental monitoring is the systematic collection, analysis, and trending of data on airborne and surface contamination within controlled manufacturing environments. In biomanufacturing cleanrooms, an environmental monitoring program provides quantitative evidence that the facility, HVAC system, personnel practices, and cleaning procedures collectively maintain the required level of particulate and microbial cleanliness for the products manufactured within those spaces.
The environmental monitoring program serves three distinct regulatory purposes. First, it demonstrates that the facility meets its classified conditions during routine operations, confirming that HVAC filtration, room pressurization, air change rates, and unidirectional airflow are functioning as designed. Second, environmental monitoring provides ongoing verification that personnel gowning, material transfer, and cleaning procedures are effective at preventing microbial ingress into classified areas. Third, the trending of environmental monitoring data over time enables detection of adverse shifts before they result in product contamination events.
A comprehensive environmental monitoring program measures both non-viable particles (using optical particle counters to detect particles at 0.5 and 5.0 micrometres) and viable organisms (using active air samplers, settle plates, contact plates, and personnel monitoring). The distinction is critical because particle counts alone cannot confirm sterility. A cleanroom may meet ISO 5 particle limits while harbouring viable organisms on surfaces if cleaning or disinfection is inadequate. Conversely, elevated particle counts during operations may reflect non-viable process-generated particles rather than microbial contamination.
For biomanufacturing facilities producing cell therapy products, monoclonal antibodies, vaccines, and other biologics, environmental monitoring is particularly important because many of these products cannot be terminally sterilized. Aseptic processing relies entirely on the controlled environment and operator technique to prevent microbial contamination of the final product. The environmental monitoring program is the primary mechanism for demonstrating that this reliance is justified.
Regulatory Framework: EU GMP Annex 1, FDA, and ISO 14644
Three regulatory frameworks govern environmental monitoring in biomanufacturing cleanrooms. Understanding how they interact and where they differ is essential for designing a program that satisfies all applicable requirements for global market access.
EU GMP Annex 1 (2022): Manufacture of Sterile Medicinal Products
The revised EU GMP Annex 1, effective 25 August 2023, represents the most comprehensive update to sterile manufacturing guidance in over a decade. For environmental monitoring, the most significant change is the mandatory Contamination Control Strategy (CCS). This must be a formal, documented strategy that identifies all potential sources of contamination, evaluates the effectiveness of each control measure, and defines the environmental monitoring approach through a risk-based methodology. The CCS is not a one-time document. It must be reviewed periodically and updated when facility, process, or procedural changes occur.
Annex 1 defines four cleanroom grades (A through D) with specific particle and microbial limits for both "at rest" and "in operation" conditions. Grade A is the critical zone for high-risk operations such as filling, making aseptic connections, and adding stoppers. Grade B is the background environment for Grade A zones. Grades C and D are for less critical steps of sterile product manufacturing such as solution preparation and equipment cleaning.
FDA Guidance: Aseptic Processing (2004) and 21 CFR 211
The FDA Guidance for Industry on Sterile Drug Products Produced by Aseptic Processing (2004) establishes the US regulatory expectations for environmental monitoring. While less prescriptive than EU GMP Annex 1 on specific frequencies, the FDA guidance requires a written environmental monitoring program with defined sampling sites, methods, and frequencies. It explicitly requires that alert and action levels be established from historical facility-specific data rather than adopted from generic industry guidance.
21 CFR 211.42 requires that aseptic processing areas maintain appropriate environmental controls, and 21 CFR 211.46 requires adequate ventilation, air filtration, and pressure differentials. The FDA typically classifies cleanrooms using ISO 14644 terminology (ISO 5, ISO 7, ISO 8) rather than the EU grade system, though the particle limits are aligned between the two frameworks.
ISO 14644-1:2015 and ISO 14644-2:2015
ISO 14644-1 defines the classification methodology for cleanrooms based on particle concentrations. It specifies the minimum number of sampling locations based on room area, the minimum sample volume per location, and the statistical evaluation criteria for classification. ISO 14644-2 provides requirements for monitoring to demonstrate ongoing compliance with classification, including the definition of a monitoring plan and the statistical approach to data evaluation.
The relationship between ISO classifications and EU GMP grades is not a direct one-to-one mapping. Grade A corresponds to ISO 5 in both at-rest and in-operation conditions, while Grade B is ISO 5 at rest but ISO 7 in operation. This distinction reflects the recognition that personnel activity generates particles, and the in-operation limits account for this contribution while still maintaining adequate cleanliness for the process.
Cleanroom Classification: Particle and Viable Limits by Grade
Cleanroom classification defines the maximum permissible particle and microbial contamination for each grade. These limits represent the regulatory specification. Your facility-specific alert and action limits must be set below these specification limits to provide early warning of process drift. The following tables present the complete EU GMP Annex 1 (2022) limits for environmental monitoring in biomanufacturing cleanrooms.
Particle Limits per Cubic Metre (EU GMP Annex 1, 2022)
| Grade | ISO Class | ≥0.5 µm At Rest | ≥0.5 µm In Operation | ≥5.0 µm At Rest | ≥5.0 µm In Operation |
|---|---|---|---|---|---|
| A | ISO 5 | 3,520 | 3,520 | 20 | 20 |
| B | ISO 5 / ISO 7 | 3,520 | 352,000 | 29 | 2,900 |
| C | ISO 7 / ISO 8 | 352,000 | 3,520,000 | 2,900 | 29,000 |
| D | ISO 8 | 3,520,000 | Not predetermined | 29,000 | Not predetermined |
Microbial Limits (EU GMP Annex 1, 2022)
| Grade | Active Air (CFU/m³) | Settle Plate 90 mm (CFU/4h) | Contact Plate 55 mm (CFU/plate) | Glove Print (CFU/glove) |
|---|---|---|---|---|
| A | <1 (zero expected) | <1 (zero expected) | <1 (zero expected) | <1 (zero expected) |
| B | 10 | 5 | 5 | 5 |
| C | 100 | 50 | 25 | - |
| D | 200 | 100 | 50 | - |
The Grade D "in operation" particle limits are listed as "not predetermined" in EU GMP Annex 1 because Grade D rooms support non-critical manufacturing steps where in-operation particle concentrations are heavily dependent on the specific process being performed. Each facility must establish its own in-operation limits for Grade D based on process characterisation data.
Cleanroom Limits Visualisation (Logarithmic Scale)
The logarithmic scale in the chart above illustrates why environmental monitoring programs must use different instrumentation sensitivity and sampling strategies for each cleanroom grade. Grade A limits span from less than 1 CFU (viable) to 3,520 particles per cubic metre (non-viable), while Grade D limits extend to 3,520,000 particles. This five-order-of-magnitude range means that a single environmental monitoring protocol cannot serve all grades effectively.
Designing Your Sampling Plan: Location Selection and Frequency
The sampling plan is the operational core of your environmental monitoring program. It defines where samples are collected, how often, by what method, and who performs the sampling. Under EU GMP Annex 1 (2022), the sampling plan must be justified within your Contamination Control Strategy based on a documented risk assessment that considers the proximity of sampling locations to exposed product, personnel activities, airflow patterns, and historical contamination data.
Location Selection Principles
Environmental monitoring sampling locations are selected using a risk-based approach that prioritizes positions where contamination would have the greatest impact on product quality. For Grade A zones, every location adjacent to exposed product, open containers, and aseptic connections must be monitored. This includes positions directly above filling needles, beside stopper bowls, and at the point of aseptic additions.
For Grade B background rooms, sampling locations should include personnel entry points (airlocks and gowning room exits), material transfer points (pass-through hatches and rapid transfer ports), equipment exhaust locations, ceiling and wall return air locations, and positions along operator traffic patterns during routine interventions. The worst-case approach means prioritizing locations where contamination is most likely to occur, not locations where the cleanest results are expected.
ISO 14644-1 Table A.1 provides the minimum number of sampling locations for particle classification based on clean zone area. For a 28-square-metre Grade B room, this requires a minimum of 7 particle monitoring locations. However, viable monitoring locations are determined by the risk assessment rather than a statistical formula. A typical Grade B filling suite of similar size may require 8 to 12 viable monitoring locations to adequately cover all identified risk points.
Sampling Frequency by Grade
| Method | Grade A | Grade B | Grade C | Grade D |
|---|---|---|---|---|
| Particle monitoring | Continuous (mandatory) | Continuous or each shift | Each shift | Twice daily |
| Viable air (active) | Each shift | Each shift | Daily | Twice per week |
| Settle plates | Continuous (4h max exposure) | Each shift | Daily | Twice per week |
| Surface monitoring | Each shift | Each shift | Weekly | Weekly to monthly |
For Grade A, the requirement for continuous particle monitoring means that particle counters must operate throughout the entire duration of all aseptic operations, with real-time alarming if counts exceed limits. Settle plates must be exposed continuously during operations, with individual plates replaced at maximum 4-hour intervals. This maximum exposure time ensures that the nutrient media remains viable for organism recovery and that overgrowth of fast-growing organisms does not mask slower-growing contaminants.
Sampling Methods Comparison
| Method | Principle | Sample Size | Advantages | Limitations |
|---|---|---|---|---|
| Active air sampling | Impaction of air onto agar surface at known flow rate | 1,000 L (typically 1 m³) | Quantitative (CFU/m³); defined volume; good recovery | Point-in-time sample; sampler may disrupt airflow; desiccation stress on organisms |
| Settle plates | Passive gravitational settling onto exposed agar | 90 mm plate, 4h exposure | Continuous monitoring; no equipment; detects large particles carrying organisms | Not quantitative per volume; influenced by airflow; limited to large particles |
| Contact plates (RODAC) | Direct surface imprint onto convex agar | 55 mm plate (25 cm²) | Direct surface assessment; reproducible contact area; quantitative | Flat surfaces only; residual media must be removed; not suitable for irregular surfaces |
| Swabs | Physical collection from surface, transfer to media | Defined area (typically 25 cm²) | Access irregular surfaces; valves, crevices; recovery factor applied | Operator-dependent; variable recovery; semi-quantitative without validated recovery factor |
The choice between sampling methods depends on the monitoring objective. Active air sampling provides quantitative data per unit volume and is required for Grade A and B viable monitoring. Settle plates complement active air by providing continuous qualitative assessment of airborne viable particles during operations. Contact plates assess surface cleanliness on flat, accessible surfaces after cleaning and disinfection. Swabs are used for hard-to-reach or irregular surfaces such as equipment joints, valve assemblies, and threaded connections.
Endotoxin Calculator
Calculate endotoxin limits for your process. Environmental monitoring programs should include endotoxin monitoring of WFI systems and Grade A surfaces.
How Do You Set Alert and Action Limits for Environmental Monitoring?
Alert and action limits are facility-specific thresholds that provide early warning of potential contamination control failures before regulatory specification limits are breached. These limits are derived from your own historical environmental monitoring data, not from textbook values or guidance documents. The regulatory specification limits (10 CFU/m³ for Grade B, 100 CFU/m³ for Grade C, 200 CFU/m³ for Grade D) represent the maximum allowable values. Your operational alert and action limits should be substantially lower.
Statistical Derivation Method
The recommended approach for setting environmental monitoring alert and action limits uses percentile analysis of historical data. Collect a minimum of 6 to 12 months of environmental monitoring data under normal operating conditions (excluding known excursions attributed to specific assignable causes). Then calculate the following:
- Alert limit: Set at the 90th percentile of the historical dataset. This means that under normal conditions, no more than 10% of results should exceed the alert limit.
- Action limit: Set at the 95th percentile of the historical dataset. Under normal conditions, no more than 5% of results should exceed the action limit.
Both limits must remain below the regulatory specification limit. If the calculated percentile values approach or exceed the specification limit, this indicates that the facility is operating too close to its maximum capability and process improvements are needed before establishing limits.
Typical Alert and Action Limits by Grade
| Grade | Specification (CFU/m³) | Typical Alert (CFU/m³) | Typical Action (CFU/m³) | Response |
|---|---|---|---|---|
| A | <1 (zero expected) | Any recovery | Any recovery | Immediate investigation; batch impact assessment |
| B | 10 | 3 | 5 | Increased monitoring; root cause investigation at action |
| C | 100 | 25 | 50 | Review cleaning; investigate at action |
| D | 200 | 50 | 100 | Review HVAC; investigate at action |
For Grade A environmental monitoring, any microbial recovery represents both an alert and an action limit excursion simultaneously. The 2022 revision of EU GMP Annex 1 reinforced the zero-tolerance approach for Grade A, explicitly stating that "the expected result is zero CFU" and that any recovery must trigger investigation. This is the most stringent requirement in pharmaceutical environmental monitoring and reflects the direct exposure of sterile product to the Grade A environment.
Worked Example: Setting Alert and Action Limits for a Grade C Cleanroom
Scenario: A Grade C solution preparation room has been operational for 14 months. You have 312 active air sampling results collected during routine environmental monitoring (twice-daily sampling, excluding weekends and planned shutdowns). The specification limit is 100 CFU/m³.
Step 1: Sort and clean data. Remove 4 results attributed to confirmed assignable causes (HVAC filter failure, door seal breach) with documented CAPAs. Working dataset: 308 results.
Step 2: Calculate percentiles.
- Sorted data: 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, ... (most values 0-8 CFU/m³)
- 90th percentile position: 0.90 x 308 = 277.2 (round up to position 278)
- Value at position 278: 18 CFU/m³
- 95th percentile position: 0.95 x 308 = 292.6 (round up to position 293)
- Value at position 293: 32 CFU/m³
Step 3: Verify against specification. Both values (18 and 32) are well below the 100 CFU/m³ specification. Adequate margin exists.
Step 4: Round to practical values.
- Alert limit: 20 CFU/m³ (rounded from 18 for practical application)
- Action limit: 35 CFU/m³ (rounded from 32 for practical application)
Step 5: Document and implement. Record the data range, exclusion rationale, calculation method, and final limits in your environmental monitoring SOP. Set review frequency at 12 months or after any significant facility change.
Microbial Trending and Statistical Process Control
Individual environmental monitoring results in isolation tell you only whether a single sample passed or failed. Trending transforms this binary assessment into a predictive tool that identifies adverse shifts in contamination control before they generate excursions. Statistical process control applied to environmental monitoring data enables you to distinguish between normal process variation (common cause) and signals indicating a genuine change in facility performance (special cause).
Control Chart Application to EM Data
The most effective trending approach for environmental monitoring data uses individuals control charts (I-MR charts) or exponentially weighted moving average (EWMA) charts. Traditional X-bar and R charts are less suitable because environmental monitoring data is typically highly skewed (most values near zero with occasional higher counts) and sample sizes per monitoring session are small.
For each monitoring location, plot the results chronologically and calculate the mean and upper control limit (UCL = mean + 3 standard deviations). Because environmental monitoring data is non-normally distributed, consider using a logarithmic transformation (log(count + 1)) before calculating control limits, or use non-parametric approaches based on the median and interquartile range.
Key trending rules adapted from Western Electric rules for environmental monitoring data include:
- One point above the action limit (immediate investigation)
- Two of three consecutive points above the alert limit (adverse trend)
- Seven consecutive points on one side of the mean (process shift)
- Seven consecutive points trending upward (drift pattern)
- Seasonal pattern exceeding historical norms for that calendar period
Example: 12-Month Viable Air Trending in Grade B
The trending chart above shows a typical pattern for a well-controlled Grade B environment. Location 1 (near the filling machine) consistently runs lower than Location 3 (near the personnel airlock), which is expected given the proximity to the contamination source. The seasonal increase visible in months 7 to 9 (summer) is a common pattern associated with increased ambient bioburden, higher humidity, and HVAC load changes. The single excursion at Location 3 in month 8 triggered an investigation that identified a worn door gasket seal at the airlock. This type of seasonal trending intelligence is only visible through sustained monthly monitoring analysis.
Monthly and Quarterly Review Cadence
Environmental monitoring trending should operate on a tiered review cadence. Daily or shift-level review catches individual excursions and triggers immediate corrective action. Monthly trending reports aggregate data by location, organism type, and method to identify gradual shifts that individual results would not reveal. Quarterly management reviews assess overall programme health, compare current performance against alert and action limits, and determine whether limit revision is warranted based on facility improvements or degradation.
Annual reviews should encompass a complete statistical reassessment of all alert and action limits, evaluation of the sampling plan for adequacy (adding or removing locations based on accumulated risk data), and assessment of trending methodologies for sensitivity. This annual review is a specific requirement of the EU GMP Annex 1 (2022) Contamination Control Strategy.
Personnel Monitoring and Gowning Qualification
Personnel are the dominant contamination source in pharmaceutical cleanrooms. Human skin sheds approximately 10 million particles per minute during normal activity, of which approximately 10% carry viable organisms. Even with full gowning ensembles, personnel contribute the majority of viable contamination detected in Grade B and higher areas. Personnel environmental monitoring provides quantitative assessment of gowning effectiveness and operator aseptic technique.
Gowning Qualification Programme
A gowning qualification programme typically requires operators to demonstrate competency through three consecutive successful gowning sessions assessed by contact plate monitoring. The assessment includes glove prints (5 fingertips of each gloved hand pressed onto TSA plates), forearm plates, chest plate, and hood area sampling. Acceptance criteria for initial gowning qualification in a Grade B environment are typically less than 5 CFU per glove print and less than 5 CFU per gown contact plate, matching the EU GMP Annex 1 specification limits for Grade B surfaces.
Requalification frequency depends on the facility quality system. Quarterly or semi-annual requalification is common, with additional requalification triggered by extended absence (typically greater than 30 days), deviation involvement, or a change in gowning procedure. Operators who fail gowning qualification must undergo retraining before re-entry to classified areas.
Routine Personnel Monitoring
Routine personnel monitoring occurs during or immediately after completion of aseptic operations. In Grade A/B environments, glove prints are taken each shift for all operators who have performed interventions. The timing is important. Glove prints taken immediately after an aseptic intervention (within the RABS or isolator) reflect actual process risk, while glove prints taken at gowning exit reflect cumulative gowning integrity over the entire shift.
Results from personnel environmental monitoring should be trended per individual operator, enabling identification of operators who consistently generate higher counts. This is not a punitive measure but a training opportunity. Operators with trending counts near alert limits may benefit from refresher training on specific techniques such as disinfection spray application, slow deliberate movements, or proper port entry technique.
Buffer Calculator
Calculate buffer compositions for WFI and purified water systems. Water system qualification underpins cleanroom environmental monitoring programmes.
Excursion Investigation and CAPA Workflow
An environmental monitoring excursion occurs when a result exceeds the defined action limit (or alert limit for Grade A). The investigation must determine whether the excursion represents a true loss of environmental control that could have impacted product quality, or whether it is an isolated event with an assignable cause that does not indicate systemic failure. The investigation outcome directly affects batch disposition decisions for any product manufactured during the excursion period.
Immediate Response (Within 24 Hours)
Upon identification of an action limit excursion, the immediate response includes notification of quality assurance, increased environmental monitoring at the affected location and adjacent locations, review of HVAC system data (differential pressures, air velocities, HEPA filter integrity), review of recent activities in the affected area (interventions, personnel changes, maintenance), and identification of all batches manufactured during the period the excursion may have been active.
Investigation Elements
A thorough environmental monitoring excursion investigation addresses the following elements systematically. Organism identification to genus and species level determines whether the recovered organism is personnel-associated (Micrococcus, Staphylococcus) suggesting a gowning or technique issue, environmental (Bacillus, moulds) suggesting HVAC or facility integrity compromise, or water-associated (Pseudomonas, Burkholderia) suggesting water system contamination.
Temporal correlation identifies whether the excursion coincides with specific events such as maintenance activities, personnel training sessions, construction in adjacent areas, seasonal weather patterns, or HVAC system modifications. Spatial correlation determines whether the excursion is isolated to a single location (suggesting a local cause) or affects multiple locations simultaneously (suggesting a systemic issue such as HVAC failure).
CAPA and Batch Impact Assessment
The corrective action depends on root cause determination. Personnel-associated organisms may require retraining, gowning requalification, or procedural changes to intervention techniques. Environmental organisms may require HVAC investigation, filter integrity testing, or facility repair. The preventive action must address the systemic vulnerability that allowed the excursion, not merely respond to the symptom.
Batch impact assessment evaluates whether any product manufactured during the period of potential environmental compromise is safe for release. For Grade A excursions, the default position is that all batches manufactured during the excursion period are held pending investigation. Release requires documented evidence that the excursion could not have resulted in product contamination, supported by sterility testing, bioburden data, and risk assessment.
All environmental monitoring excursion investigations and their outcomes should be reviewed during periodic quality management reviews and trending analysis. Recurring excursions at the same location, with the same organism, or during the same seasonal period indicate inadequate preventive action and require escalation to facility or programme-level corrective actions.
Frequently Asked Questions
How often should you perform environmental monitoring in a Grade A cleanroom?
Grade A cleanrooms require continuous particle monitoring throughout all aseptic operations, with viable air sampling each shift, settle plates exposed continuously for a maximum of 4 hours per plate, and surface monitoring (contact plates and glove prints) performed each shift. EU GMP Annex 1 (2022) mandates that any microbial recovery in Grade A is treated as an action limit breach with immediate investigation, since the expected result is zero CFU at all times. Continuous particle counters must alarm in real time if the 3,520 particles per cubic metre limit (at 0.5 micrometres or larger) is exceeded.
What is the difference between alert and action limits in EM?
Alert limits are early-warning thresholds set below specification limits, typically at the 90th percentile of historical data. Exceeding an alert limit triggers increased monitoring and investigation to determine if the process is drifting out of control, but does not necessarily require stopping production. Action limits are set at the 95th percentile of historical data, closer to but still below the regulatory specification limit. Exceeding an action limit requires immediate corrective action, which may include stopping operations, environmental decontamination, and a formal CAPA investigation. Both limits are site-specific and derived from at least 6 to 12 months of facility-specific environmental monitoring data.
What organisms are most commonly recovered in cleanroom environmental monitoring?
The most commonly recovered organisms in pharmaceutical cleanrooms are human-associated microflora shed from personnel during gowning and operations. Micrococcus species, coagulase-negative Staphylococcus (particularly S. epidermidis and S. hominis), and Corynebacterium species dominate cleanroom isolates, typically accounting for 70 to 90 percent of all recoveries. Environmental organisms such as Bacillus species and moulds (Aspergillus, Penicillium, Cladosporium) indicate potential HVAC or facility integrity issues. Recovery of Gram-negative organisms (Pseudomonas, Burkholderia) is particularly concerning as these suggest water system contamination. All isolates should be identified to genus level minimum, with species-level identification for action limit excursions.
How many sampling locations are needed for a cleanroom EM program?
The number of environmental monitoring sampling locations depends on room size, classification grade, and risk assessment. ISO 14644-1 Table A.1 provides minimum sampling locations for particle classification based on room area (for example, 9 locations for a room up to 52 square metres at ISO 5). For routine viable monitoring, sampling locations are determined by risk assessment per EU GMP Annex 1 (2022) Contamination Control Strategy requirements. As a practical guideline, Grade A zones need monitoring at every critical point adjacent to exposed product, Grade B rooms typically require 6 to 12 locations per room depending on size and activities, and Grade C/D areas require proportionally fewer locations based on assessed risk. Locations must include worst-case positions: near doors, personnel traffic paths, equipment exhaust, and areas with reduced airflow.
What changed in environmental monitoring requirements with EU GMP Annex 1 (2022)?
The revised EU GMP Annex 1 (effective August 2023) introduced several significant changes to environmental monitoring requirements. The most impactful is the mandatory Contamination Control Strategy (CCS), a formal documented strategy that must identify all contamination risks, critical control points, monitoring locations, frequencies, and methods through a risk-based approach. Other key changes include: continuous particle monitoring in Grade A is now explicitly mandatory (not just recommended); the concept of corrective actions for Grade A has been strengthened to zero tolerance for any viable recovery; settle plate exposure time was clarified as a maximum of 4 hours; and there is increased emphasis on trending, with requirements to analyse environmental monitoring data for adverse trends rather than just individual excursions. The CCS must be a living document subject to periodic review and update.
References
- EU GMP Annex 1 (2022). Manufacture of Sterile Medicinal Products. European Commission Health and Consumers Directorate-General.
- ISO 14644-1:2015. Cleanrooms and associated controlled environments. Part 1: Classification of air cleanliness by particle concentration. International Organization for Standardization.
- USP <1116> Microbiological Evaluation of Clean Rooms and Other Controlled Environments. United States Pharmacopeia.
- FDA Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing. Current Good Manufacturing Practice (2004). U.S. Food and Drug Administration.
- PDA Technical Report No. 13 (Revised 2001, Errata 2021). Fundamentals of an Environmental Monitoring Program. Parenteral Drug Association.