270 BIOLOGICAL RADIATIONS AND SPECIATION Figure Speciation and the gradualist model The present day radiolarian species Pterocanium charybdeum extends well back into the Miocene where a subspecies (P c allium is recognized) At about million years ago a new species, P prismatium arose and subsequently went extinct about 1.8 million years ago To compile this diagram samples representing populations at 50 000 year intervals were taken from piston cores during the key 4.5 3.5 million year interval (much more widely spaced outside of this time band) The radiolarian tests were measured for 32 morphological characters such as length, breadth, angles and outlines and subjected to a multivariate discriminant analysis The x axis of the diagram shows the value of the discriminant function that best separates the species The boxes are the population means Ỉ standard error The histograms on the right show discriminate scores at the point of the intermediate population (lower diagram) and the sampling level 50 000 years later (upper diagram), showing clear separation at this time; these two documenting the speciation event Notice that both lineages continue to diverge morphologically in a gradualist manner long after the speciation event Reproduced from Lazarus DB (2001) Speciation and morphological evolution In: Briggs DE and Crowther PR (eds.) Palaeobiology II, pp 133 137 Oxford: Blackwell Scientific Interactions between organisms – such as the extinction of one kind allowing ecological and phenotypic diversification in another Alternatively it may be true that the morphological and diversification of one group allows similar phenomena in another (e.g., insects and plants) For example, in fishes, it has long been noted that the evolution of many durophagous (mollusc crushing) lineages coincides with the Mesozoic diversification of bivalve and gastropod molluscs Many of these ‘associations’ are anecdotal and, while they may well be true, they need firm experimental evidence for their justification It is rarely possible to isolate a single cause and in all probability complex interactions between two or more factors are ultimately responsible for radiations Therefore, the study of these evolutionary phenomena is both frustrating and challenging, demanding lengthy and detailed data collection Factors which may distort our view of radiations include the following: Imperfections of the fossil record Large hiatuses or differential preservation will inevitably distort our perception of radiations For example, the sudden appearance and apparent radiation of many catfishes (most of which are freshwater) in the Early Tertiary must be judged against the knowledge that there are very few freshwater fish-bearing deposits in the underlying Upper Cretaceous Appearance of many diverse animals and plants associated with Lagerstaătten deposits Such deposits dramatically increase the numbers and morphological diversity of species that may, in reality, have had a long unrecorded history