In the Random Island area on the west side of Trinity Bay, eastern Newfoundland, there are three NNE-SSW oriented Lower Palaeozoic basins. The southern part of the westernmost basin underlies the northern end of Random Island; it includes, apart from minor fault-bounded remnants of Lower Cambrian rocks, an almost complete, succession of Middle Cambrian to lowermost Ordovician (Tremadoc) rocks. Outcrops are largely confined to coastal cliffs. -- Two sections were sampled for acritarchs. The first, on the west side of the island in rocks of Middle Cambrian to early Tremadoc age, extends from Cock and Hen Point northward for 6.5 km to an unnamed point 2.5 km beyond the village of Elliott's Cove. The second, on the east side of the island, is a short section (a few hundred metres in length) in Upper Cambrian rocks 2 km north of Snooks Harbour. The Upper Cambrian rocks of this section complement those present in the section north of Cock and Hen Point where the former are unrepresented due to faulting. Descriptions are given of 102 species, belonging to 33 genera; 38 species and one genus are new finds. A second new genus represents a reassignment of a previously-known species. -- The acritarch microfloras of the Middle Cambrian to Tremadoc rocks on Random Island have been divided, on the basis of the successive first appearances of distinctive species, into a series of nine acritarch assemblages, RI 1 to RI 9. Two of the assemblages, RI 6 and RI 7, in rocks of late Upper Cambrian age, have not previously been found elsewhere, and assemblage RI 8, in rocks of latest Upper Cambrian age has until now been recorded only from the Obolus Beds of the northwestern part of the Russian Platform and the Obolus Zone of northern Norway. These Obolus-bearing beds, although presently included in the Tremadoc are shown to be of latest Upper Cambrian age. -- The Middle Cambrian to Tremadoc acritarch assemblages of Random Island are compared with those that have been distinguished in rocks of comparable age in England and Wales, Belgium, Spain, northern Norway, the Russian Platform, North Africa, and elsewhere. This comparison shows that those successions are incomplete due to stratigraphic breaks. Furthermore, many of the acritarch assemblages (or zones) erected elsewhere have not been accurately dated using established trilobite zones; most of the Random Island acritarch assemblages have been directly related to such zones. Thus, outside Random island, none of the sequences of acritarch assemblages (or zones) established for Middle Cambrian to Tremadoc rocks is as complete, or has as many of its assemblages as accurately dated, as that described in this study. The succession of acritarch assemblages on Random Island should, therefore, become the standard as far as rocks of Middle Cambrian to early Tremadoc age are concerned for purposes of biostratigraphic correlation.