The Eastern Hindukush forms part of an elongate belt ("Central Domain", collage of Cimmerian microcontinents) that encircles the northern part of the Indian subcontinent. A Gondwanan origin is commonly assumed for this belt, but a "Laurasian" origin for the Chitral region has been argued on palaeontological (Talent and Mawson, 1979) and palaeomagnetic (Klootwijk and Conaghan, 1979) grounds. The "Laurasian" view was based on a pilot study we undertook of Upper Devonian pisolitic ironstones from a thrust sheet at Kurāgh Spur in Chitral. Preliminary results showed a characteristic magnetization component [D = 318°, I= -6.5°, N = 7 (block samples), k = 14, α95 = 16.5°] indicating an equatorial palaeoposition. This component was thought to be of primary origin and was interpreted in terms of a Late Devonian "Laurasian" affinity of the Kurāgh Spur rocks. This controversial conclusion has been tested in the present more comprehensive study of the thrust pile of sedimentary rocks in the Reshūn-Kurāgh-Būni region of Chitral and the primary origin of the characteristic magnetization component refuted. Thermal demagnetization of 333 block samples from Middle to Upper Devonian variegated sediments, Permian quartz flysch, Permo-Triassic carbonates, and mid-Cretaceous redbeds showed two interpretable components. A softer component of recent origin (A); and a harder characteristic component (B) of both normal and reverse polarity whose mean direction [D = 314.1°, I = 6.0°, N = 4 (thrust sheets), k = 198.2, α95 = 6.5°] is comparable to the characteristic component observed in our preliminary study. However, the universal presence of this component throughout the thrust pile proves its overprint origin, which we attribute to initial India-Asia contact. Palaeomagnetic information pertinent to the controversy of a "Laurasian" versus a Gondwanan origin of the Chitral region has not been obtained in this further study because primary magnetizations could not be identified beyond doubt. Hence, we retract herewith our original conclusion of a Late Devonian "Laurasian" affinity of the Chitral region on the basis of the palaeomagnetic evidence. The secondary component (B) comprises a suite of secondary magnetizations, acquired at equatorial-to-low-northern palaeolatitudes, and is attributed to initial contact between Greater India and southern Asia. Component B has been observed previously in the Himalayan-Tibetan region, both north and south of the Indus-Tsangpo Suture zone. Identification is herein extended to the Hindukush region north of the Northern Kohistan (or Shyok) Suture zone, which is a western continuation of the Indus-Tsangpo Suture. Comparison of this suite of collision-at-tributed equatorial palaeolatitude data from the India-Asia convergence zone with new palaeolatitude constraints from the Ninetyeast Ridge on the northward movement of the Indian plate, constrained additionally by a recent minimal estimate of the palaeogeographic northern extent of Greater India, indicates that initial contact between northwestern Greater India and southern Asia was established at, or before, the Cretaceous-Tertiary boundary. The overprint origin of component B at about this time is further supported by observations by Zeitler (1985) on rocks from the sampled area in Chitral of partially reset zircon fission-track ages around 68-55 Ma. The NW-SE declination axis of component B indicates a 60-70° counterclockwise rotation of the sampled thrust pile with respect to Eurasia and a counterclockwise rotation between 10 and 30° with respect to India. Some of the recent field components (A) show a comparable rotation and indicate that the tectonic activity that led to the formation of the Hindukush-Pamir-Karakorum syntaxial zone has continued into recent times.