Multi-analytical approaches, combining the use of non-invasive and micro-invasive techniques, are increasingly being applied to the study of historical and archaeological textile collections, with particular attention to the technology of textile production, dye stuff sources and dyeing practices. By contrast, although the use of multispectral imaging (MSI) techniques is well-established in the study of polychrome surfaces, these have only been sparingly, and often unsystematically, applied to the investigation of textiles. This work summarises some of the research recently undertaken at the British Museum that applied MSI techniques to the investigation of two groups of archaeological textiles from the British Museum’s collections: Some examples of Late Antique (c. 250-800 AD) textiles from Antinoupolis and Wadi Sarga, Egypt, and a series of textile fragments from Dunhuang,China, dated from the late 7th to the early 10th century (Tang to Five Dynasties period). The aim is to show how this non-invasive, relatively inexpensive, portable methodology can be used to map the photoluminescence and reflective characteristics of textiles under different wavelengths of light. Standardised acquisition and post-processing methods were applied to produce a series of images that provided preliminary indications of the colourants used and their spatial distribution. To assess the potential and limitations of relating multispectral data to chemical properties, the information derived from these images was compared to the more detailed information provided by complementary non-invasive techniques, such as fibre optic reflectance spectroscopy (FORS), and micro-invasive approaches, such as high-performance liquid chromatography mass spectrometry (HPLC-MS). Two salient points emerge: MSI techniques show a high degree of potential for the investigation of textiles in museum/historic collections, particularly as part of a protocol combining digital microscopy, FORS, SEM-EDX and HPLC-MS. Secondly, that within this protocol, MSI is an excellent aid in planning more targeted and effective sampling strategies and facilitates comparisons between objects.