The ways in which female mammals produce offspring are shaped by natural selection: the genes of mothers that produce offspring that are too small to survive or too large or numerous to be successfully reared, disappear from a population. And yet many different ways of producing offspring have evolved in mammals. Adult female short-tailed shrews (Blarina brevicauda) and big brown bats (Eptesicus fuscus) both weight about 18 grams, yet the bat has a gestation twice as long as that of the shrew's, and gives birth to young that are more than three times heavier. The shrew, on the other hand, produces litters of up to nine offspring, whilst the bat gives birth to singletons. These differences lie at opposite ends of a continuum of reproductive strategies. How and why did such differences evolve? How do they fit together into an overall 'architecture' of maternal reproductive investment?
We aim to answer these questions through the most detailed and comprehensive comparative study of mammalian female reproduction and offspring development yet undertaken, with particular attention for those traits that are still little understood. For example, we are testing hypotheses on the fascinating variation in the morphology of the placenta in different mammals and the implications this has for offspring development. We are also examining how prenatal traits (such as gestation length and neonatal encephalization) relate to offspring development and postnatal characteristics (such as age and mass at weaning). Finally, we will test hypotheses on the coevolution of placental morphological diversity and disease load.
We are building a large database of reproductive and developmental characteristics, such as the type of placenta,the composition of the milk, the size and encephalization of newborn offspring, and using this database together with an array of phylogenetic comparative methods to study the evolution of reproductive strategies.
Our data will be made available to other scientists via an online database, providing both information on individual species, and a resource for future comparative studies.
Evolution of the placenta
The placenta is responsible for the transfer of nutrients from mother to the foetus and waste products in the opposite direction. Despite undertaking these functions in all species, the placenta is amongst the most morphologically variable of mammalian organs. This variation has remained an enigma as, despite much speculation concerning its implications for maternal investment, foetal growth and parent-offspring conflict, no such implications have previously been demonstrated. We show that the degree of interdigitation of the placenta is associated with gestation length and growth rates. When the placenta has a highly interdigitated (labyrinthine) structure, gestation length is about half as long as when the placenta is less interdigitated (villous or trabecular). Because there is no difference in brain or body size of the newborn, foetal growth rates must be correspondingly higher in species with labyrinthine placentas. Previously it had been suggested that a different aspect of placental structure, invasiveness - the degree foetal tissues erode maternal tissues to gain direct contact with the maternal blood supply - might correlate with foetal growth. However we show that, after accounting for interdigitation, invasiveness is uncorrelated with both gestation length and neonate size. We suggest that interdigitation, and the tradeoffs between growth rates and gestation length, should be incorporated in theoretical models on the evolutionary dynamics of parent-offspring conflict (see Capellini et al. 2011 Am. Nat.).
Evolutionary development of brain size
Brain size variation in mammals is associated with life histories: larger-brained species have longer gestations, mature later and have increased lifespans. Explanations for these patterns have been framed in terms of both developmental costs (larger brains take longer to grow) and cognitive benefits (larger brains enhance survival and increase lifespan). In support of the developmental costs hypothesis, we show that evolutionary changes in pre- and post-natal brain growth are associated specifically with the duration of the relevant phases of maternal investment (gestation and lactation respectively). We also find support for the hypothesis that the rate of foetal brain growth is related to the energy turnover of the mother. In contrast, we find no support for hypotheses proposing that developmental costs of brain growth are accommodated through trade-offs between brain and body growth, or between brain growth and litter size. Once the duration of maternal investment is taken into account, however, adult brain size is unrelated to post-weaning life history traits and lifespan. Hence, the genral pattern of slower life histories in larger-brained mammals appears to be a direct consequence of developmental costs (see Barton & Capellini 2011, PNAS).
Brains, genes and development
As part of this project we are collaborating with Nick Mundy and Steve Montgomery, molecular geneticists at Cambridge University, to investigate the correlated evolution of genes and brain development. We showed that brain size increased independently in many primate lineages but also that decreases occurred in a few lineages (e.g. mouse lemurs and marmosets; Montgomery et al. 2010 BMC Biol.). Moreover, we found that the decrease observed in Homo floresiensis is not unusual when evaluated in the light of the whole macroevolutionary pattern in Primates (Montgomery et al. 2010 BMC Biol.). We have also studied the genetic correlates of neonatal and adult brain size in order to determine how genes and neuro-development are linked. We have found that four microcephaly genes important in brain development are under positive selection in Primates, but only two of them (ASPM and CDK5RAP2) are associated with phenotypic increases in neonatal brain size, probably through their effect on prenatal neural proliferation. These results indicate that primate brain size has a partially conserved genetic basis (Montgomery et al. 2011 Mol. Biol. Evol.).
Metabolism, size and reproduction
Metabolic rates are believed to reflect important physiological constraints on reproduction, life history and ecology. Such constraints are often inferred from similarities between the allometric scaling of metabolic rates and other traits. For example, 'Kleiber's Law' specifies that metabolic rates increase with body mass to the 3/4 power, and so does the scaling of brain size to body size. Thus, it has been suggested that the 3/4 brain scaling exponent reflects a metabolic constraint on maternal investment in the growth of the offspring's brain. The 3/4 exponent has also been held to have major macro-ecological consequences. Our analyses, in collaboration with Chris Venditti at Reading University, suggest, however, that there is no general scaling exponent underlying variation in mammalian basal or total metabolic rates: metabolic allometry varies significantly across lineages and metabolic states (Capellini et al. 2010 Ecology). Our results thus undermine general life history and ecological theories based on any universal or predominant scaling exponent. We are now examining the consequences of metabolic rate variation for a range of reproductive and developmental traits.