Traditionally, boron (B) isotope ratios have been determined using thermal ionization mass spectrometry (TIMS) and, to some extent, secondary ion mass spectrometry (SIMS). Both TIMS and SIMS use a high-resolution mass analyzer, but differ in analyte ionization methods. TIMS uses electrons from a hot filament, whereas SIMS employs an energetic primary ion beam of Ga+, Cs+, or O- for analyte ionization. TIMS can be used in negative or positive ion modes with high sensitivity and precision of B isotope ratio determination. However, isobaric interferences may be a problem, if the sample is not well purified and/or memory of the previous sample is not removed. Time-consuming sample preparation, analyte (B) purification, and sample determination processes limit the applications of TIMS for routine analyses. SIMS can determine B and its isotope ratio in intact solid samples without destroying them, but has poorer resolution and sensitivity than TIMS, and is difficult to standardize for biological samples. Development of plasma-source mass spectrometry (MS) enabled the determination of B concentration and isotope ratio without requiring sample purification. Commonly used plasma-source MS uses an Ar inductively coupled plasma (ICP) as an ionization device interfaced to a low-resolution quadrupole mass analyzer. The quadrupole ICP-MS is less precise than TIMS and SIMS, but is a popular method for B isotope ratio determination because of its speed and convenience. B determination by ICP-MS suffers no spectroscopic interferences. However, sample matrices, memory effects, and some instrument parameters may affect the accuracy and precision of B isotope ratio determination if adequate precautions are not taken. New generations of plasma-source MS instruments using high-resolution mass analyzers provide better sensitivity and precision than the currently used quadrupole ICP-MS. Because of the convenience and high sample throughput, the high-resolution ICP-MS is expected to be the method of choice for B isotope ratio determination. The current state of instrumental capabilities is adequate for B isotope determination. However, precision and accuracy are primarily limited by sample preparation, introduction, and analytical methodology, including 1. Analyte loss and isotope fractionation during sample preparation. 2. The precision of B isotope determination in small samples, especially those containing low concentrations. 3. Difficult matrices. 4. Memory effects. Sample preparation by alkali fusion allows rapid and complete decomposition of hard-to-digest samples, but high-salt environments of the fused materials require extensive sample purification for B ratio determination. The alternative wet-ashing sample decomposition with HF also results in B loss and isotopic fractionation owing to the high volatility of BF3. Open-vessel dry- or wet-ashing methods usually do not work well for animal samples, and are also prone to B loss and contamination. Closed-vessel microwave digestion overcomes these problems, but the digests of biological materials have high C contents, which cause spectral interference on 11B and affect 11B/10B ratios. Exchange separation/preconcentration of B using exchange (cation or anion exchange, B-specific resin, e.g., Amberlite IRA-743) tend to cause B isotope fractionation, and C eluting from these resin columns may interfere with B isotope ratio determination. Memory effects of B that occur during sample determination may cause serious errors in B isotope ratio determination, especially when samples varying in B concentrations and/or isotope composition are analyzed together. Although the utilization of high-resolution plasma-source MS will undoubtedly improve analytical precision, it is the sample preparation, sample introduction, and analytical methodology that represent the primary limitation to accurate and precise B isotope ratio determination.