Why do females of many species mate with males on the basis of traits apparently detrimental to male survival? The answer may lie in the fact that these male traits are correlated with male condition. We consider the argument that high male condition directly benefits female fecundity and/or viability (e.g. through lower transmission of parasites, improved control of resources, or better paternal care). Using a quantitative genetic model we show how female preferences for male traits that indicate condition can evolve, even if the male traits themselves have deleterious effects on both the male and the female's fecundity. So-called 'arbitrary preferences' can spread in this way because male traits subject to sexual selection are often under additional selection to become correlated with condition. At equilibrium the positive effects of male condition on a female's fecundity and the negative effects of the male trait on her fecundity are balanced and the female preference is under stabilizing selection. The male trait will often be correlated with viability, but not with fecundity, even though the preference evolved as a rrsult of differences in male fecundity. The mean fecundity of females is not maximized, and can steadily decline as the male trait and female preference evolve. If the male trait has no direct deleterious effects on female fecundity, as may happen in species with no paternal care, female preferences are under continuous directional selection to increase. ADDITIONAL KEY WORDS:-Fecundityselection --handicapsmean fitnessfemale preferencesconditionpaternal carequantitative geneticstheoretical modelsbirds.

Human plasma fibronectin and a series of its large proteolytic fragments were analyzed by electron microscopy using tungsten shadowing on carbon and polystyrene films. On carbon, intact fibronectin appeared as a randomly coiled strand, while on polystyrene it appeared as an elongated structure. Two fragments of fibronectin, M, = 205000 and 190000, which lack the NH2-terminal domain of fibronectin and retain the collagen-binding, cellattachment and heparin-binding functions, and a M , = 170000 fragment, which retains the collagen-binding and cell-attachment functions, were seen as rods with varying degrees of nodularity while a M , = 100000 fragment, which only binds to collagen, had two clear-cut domains.These results support the existing biochemical evidence that the segregation of the functional activities in the fibronectin molecule is based on distinct structural domains and provides evidence for the existence of an additional structural domain not revealed by biochemical and functional studies.

No abstract

Surfaces of amorphous carbon, polystyrene, and polycarbonate were exposed to solutions of fibrinogen, a modified fibrinogen which lacked the alpha chain protuberance, and serum albumin. The results were studied by electron microscopy. The exposures occurred in a well characterized flow environment at a shear rate of 135/s. All three protein species formed a film when adsorbed to carbon films. When the proteins were adsorbed to polystyrene surfaces, formation of a network was observed. Polycarbonate surfaces adsorbed the proteins both as a network and as a continuous film. It was observed that the nature of the adsorption process depended upon the specific combination of molecule and material. For example, on carbon, individual fibrinogen molecules retain their trinodular structure and adsorb randomly until a monolayer forms. On polystyrene, the individual fibrinogen molecules appear as globules and a network forms before complete coverage occurs.