A large body of paleontological literature concerns the importance of ontogeny as a source of morphological variation for evolution; morphologies that appear during one stage of an organism's development are made available for use in another simply by modifying the developmental program. Paleontologists need to know why this occurs, so they can study the process of evolution in extinct animals and so they can discuss the fossil record in terms that are applicable to modern forms. If most cases of heterochrony can be attributed to life-history evolution then the fossil record provides evidence of the nature of selection (in particular the age-specific mortality) that extinct animals experienced. The hypothesis of interest here is that species in which maturity is accelerated will also show generalized morphology and small size, while those with delayed maturity will have more specialized morphology and large size.

Four species of the ostracode genus Cyprideis were studied to determine whether differences in age at maturity are correlated with heterochrony in the expected manner. For each species the changes in size and shape through geological time were evaluated in the statistical context of modern geographic and seasonal variation. Living populations were sampled regularly to detect differences in seasonality and to estimate the duration of development.

Evolution of ontogeny is apparent at the level of species in this group, but it is not simply related to differences in life-history. In comparisons among species, we find evidence of heterochrony where there is no difference in the age at maturity, and a difference in age at maturity where there is no heterochrony. Similarly, three of the four species show the expected positive correlation between size and age at maturity, yet the fourth species is relatively large and matures rapidly. Cyprideis does not support the generalization that life-history evolution causes heterochrony, and casts doubt on the inference of life-history evolution from heterochrony where the data are drawn exclusively from extinct forms.

Geographic range appears to be an important aspect of the biology of species, but ranges cannot be unambiguously determined from the fossil record: ancient species can rarely be traced to the full extent of their original geographic ranges, and some ancient provinces are far better characterized than others. Here we test the degree to which paleogeographic range data for species within ancient provinces can be used to estimate the relative magnitudes of total geographic ranges, focusing on the 212 living bivalve species of the Oregonian Province, a north-south (N-S) province of the northeastern Pacific shelf.

We break the total range of each species into its within-province and extraprovincial components. Nonparametric rank correlation tests and simple linear regressions yield highly significant correlations, both between within-province and total ranges (which are not independent variables) and between within-province and extraprovincial ranges. The strongest correlations are obtained when north-ranging species are excluded. These north-ranging species are thermally tolerant and have access to broad east-west (E-W) provinces, thus weakening the relation between their within-province and total ranges. On a coarser scale, we also found that the extraprovincial ranges of species with small within-province ranges (<650 km, ca. 25% of focal province) are significantly less than those for species with large within-province ranges (> 1950 km, ca. 75% of focal province). The significant correlations of within-province range with both total and extraprovincial ranges hold for species with within-province ranges >650 km, so caution is needed regarding species that extend into a focal province only a short distance from its edges. We conclude that within-province geographic range can generally be used to rank species by total geographic range and thus is an adequate proxy in most comparative studies of the paleobiologic consequences of range magnitudes.

Attempts at understanding evolutionary relationships among Paleozoic Dipnoi (lungfish) using cladistic methodology have proved totally unsatisfactory (Miles 1977; Marshall 1987). We attempt to reconstruct the relationships between the better known genera using a method that involves the recognition of lineages based on evolving functional complexes, particularly those involved with food reduction and respiration. Within these broadly defined lineages, we have defined sub-lineages based on evolutionary patterns shown by structures that have been stratigraphically dated; such patterns are found inter alia in the roofing bones and the external dermal bones of the mandible. A number of new suborders and families are recognised; genera for which further morphological data are required before they can be assigned to a higher taxon are indicated; two generic synonyms are recognised.

In appendices, short descriptions are given of two new genera—Pillararhynchus from the Gogo Formation (Upper Devonian) of Western Australia, and Sorbitorhynchus from the Emsian of Guangxi, China.

Living crocodilians and limbed lepidosaurs have a large caudofemoralis longus muscle passing from tail to femur. Anatomical and electromyographic data support the conclusion that the caudofemoralis is the principal femoral retractor and thus serves as the primary propulsive muscle of the hind limb. Osteological evidence of both origin and insertion indicates that a substantial caudofemoralis longus was present in archosaurs primitively and was retained in the clades Dinosauria and Theropoda. Derived theropods (e.g., ornithomimids, deinonychosaurs, Archaeopteryx and birds) exhibit features that indicate a reduction in caudofemoral musculature, including fewer caudal vertebrae, diminished caudal transverse processes, distal specialization of the tail, and loss of the fourth trochanter. This trend culminates in ornithurine birds, which have greatly reduced tails and either have a minute caudofemoralis longus or lack the muscle entirely.

As derived theropod dinosaurs, birds represent the best living model for reconstructing extinct nonavian theropods. Bipedal, digitigrade locomotion on fully erect limbs is an avian feature inherited from theropod ancestors. However, the primitive saurian mechanisms of balancing the body (with a large tail) and retracting the limb (with the caudofemoralis longus) were abandoned in the course of avian evolution. This strongly suggests that details of the orientation (subhorizontal femur) and movement (primarily knee flexion) of the hind limb in extant birds are more properly viewed as derived, uniquely avian conditions, rather than as retentions of an ancestral dinosaurian pattern. Although many characters often associated with extant birds appeared much earlier in theropod evolution, reconstructing the locomotion of all theropods as completely birdlike ignores a wealth of differences that characterize birds.

Paleozoic and post-Paleozoic marine faunas are strikingly different in composition. Paleozoic marine gastropods may be divided into archaic and modern groups based on taxonomic composition, ecological role, and morphology. Paleozoic assemblages were dominated by pleurotomariids (Eotomariidae and Phymatopleuridae), the Pseudozygopleuridae, and, to a lesser extent, the Euomphalidae, while Triassic assemblages were dominated by the Trochiina, Amberleyacea, and new groups of Loxonematoidea and Pleurotomariina. Several new groups of caenogastropods appeared as well. Yet the importance of the end-Permian mass extinction in generating these changes has been questioned. As part of a study of the diversity history of upper Paleozoic and Triassic gastropods, to test the extent to which taxonomic and morphologic trends established in the late Paleozoic are continued after the extinction, and to determine the patterns of selectivity operating during the extinction, I assembled generic and morphologic diversity data for 396 genera in 75 families from the Famennian through the Norian stages. Within this interval, gastropod genera underwent an adaptive radiation during the Visean and Namurian, largely of pleurotomariids, a subsequent period of dynamic stability through the Leonardian, a broad-based decline during the end-Permian mass extinction, and a two-phase post-extinction rebound during the Triassic. The patterns of generic diversity within superfamily-level clades were analyzed using Q-mode factor analysis and detrended correspondence analysis.

The results demonstrate that taxonomic affinity, previous clade history, generic age, and gross morphology did not determine survival probability of genera during the end-Permian extinction, with the exception of the bellerophontids, nor did increasing diversity within clades or expansion of particular morphologies prior to the extinction facilitate survival during the extinction or success after it. The pleurotomariids diversified during the Lower Permian, but were heavily hit by the extinction. Similarly, trochiform and turriculate morphologies, among those which Vermeij (1987) has identified as having increased predation resistance, were expanding in the late Paleozoic, but suffered similar extinction rates to other nondiversifying clades. Survival was a consequence of broad geographic and environmental distribution, as was the case during background periods.

The exceptionally preserved Triassic (Anisian) Grès à Voltzia fauna from the Vosges of northeastern France is compared with four major Carboniferous Konservat-Lagerstätten. The problem of comparing faunas with different types of preservation and degrees of taxonomic determination is addressed with a newly devised similarity coefficient. This coefficient quantifies and combines data from different levels in the taxonomic hierarchy and allows comparisons between biotas in a consistent direction regardless of relative diversity. The rank order of similarity between the four Carboniferous Konservat-Lagerstätten and the Grès à Voltzia is as follows: 1. Mazon Creek (Westphalian D) of Illinois; 2. Glencartholm (Visean) of southern Scotland; 3. Bear Gulch (Namurian) of Montana; and 4. Blanzy-Montceau (Stephanian) of France. These occurrences represent conditions transitional between nearshore fully marine and fresh water. The Grès à Voltzia fauna is significantly closer to the Mazon Creek fauna than to the others; the taxonomic overlap with Blanzy-Montceau and Bear Gulch is limited.

Stratigraphic age has an insignificant influence on the result; indeed, the fauna closest in age to the Grès à Voltzia, that of Blanzy-Montceau, is least similar. Taphonomic factors are important in determining the range of organisms preserved. The Glencartholm fauna is represented only by forms with either mineralized or robust chitinous skeletons, implying a greater degree of decay prior to the onset of diagenetic mineralization than in the other Konservat-Lagerstätten. Environment, however, is the major control on similarity. The Mazon Creek biota, like that of Grès à Voltzia, represents settings transitional between terrestrial and marine-influenced delta. Groups common to both Konservat-Lagerstätten include medusae, brachiopods, polychaetes, bivalve and gastropod molluscs, limulids, scorpions, spiders, branchiopods, ostracodes, malacostracans, cycloids, euthycarcinoid and myriapod uniramians, insects, fish, and tetrapods. There is a striking continuity between the faunas of Carboniferous and Triassic transitional sedimentary environments. Groups that were adapted to fluctuating conditions (e.g., shifting salinity) show strong congruence at the family and lower levels and were little affected by Permian extinctions. The major taxonomic contrasts are in the eumalacostracans and insects: many of the groups represented in the Grès à Voltzia appeared in the Permian and radiated across the Permo-Triassic boundary as Paleozoic forms became extinct.

Amino acid analyses of undecalcified samples of fossil crocodile and rhinoceros enamel and dentin from mature teeth revealed that the total protein content of these mineralized fossil tissues varied from ~0.01–0.007% by weight. Except in one instance, amino acid analyses of the enamel proteins revealed them to be free of collagen and to have an amino acid composition similar to the proteins obtained from the enamel of mature modern vertebrates. Molecular sieving of the acid soluble enamel proteins demonstrated that the components consisted principally of small peptides and free amino acids, as in the enamel of modern vertebrates.

Based on the presence of hydroxyproline (hyp) and hydroxylysine (hyl), collagen was detected in undecalcified mature dentin of fossil rhinoceros, but not in undecalcified crocodile dentin. It was only by sequential extraction procedures that the presence of collagen in the dentin of fossil crocodile was established, emphasizing the utility and importance of analyzing the soluble components in specific extracts of the fossil. Based on these data and the concentrations of hyp and hyl in the material solubilized by the various solvents, the dentin of fossil rhinoceros contained considerably more collagen per weight and as a percentage of the total protein in the dentin than did the dentin of fossil crocodile.

As with modern dentin, EDTA and dilute acid solubilize the noncollagenous proteins and peptides found in fossil enamel and dentin, some of which contain O-phosphoserine [Ser(P)], an amino acid unique to mineralized connective tissues. Similar to recent reported findings from fossil bone, less of the original content of the noncollagenous proteins, including those phosphorylated, is degraded and removed from the enamel and dentin during fossilization than the percentage of dentinal collagen and the nonphosphorylated domains of the enamel proteins which are removed. This selective resistance to degradation and removal of the noncollagenous proteins, including phosphoproteins, may reflect the strong interaction of these proteins with the mineral phase of fossilized tissues. The amount and close packing of the inorganic crystals may also inhibit the interaction of the proteins in the interior of the enamel and dentin with the geochemical environment.

This short contribution clarifies similarities and differences between the findings of Gilinsky and Bambach (1986) and Gilinsky and Good (1989) regarding the shapes of clades. The methods developed in the latter contribution are preferred.