Annu. Rev. Astron. Astrophys. 1988. 26: 561-630 Copyright © 1988 by Annual Reviews. All rights reserved

9.2. The Value of T₀ From the Chemical Elements and From the Oldest Stars

To use Figure 14 requires a knowledge of H₀ and the age of the Universe, found independently of the expansion data. The two methods to measure T₀ are (a) use of the age of the oldest stars in the Galaxy (to which the gestation time of galaxies is added), and (b) a determination of the age of the chemical elements. Entrance to the large literature on the second problem is available in Burbidge et al. (1957), Dicke (1969), Cameron (1982), Fowler (1984), Thielemann & Truran (1986), Butcher (1987), Clayton (1987), and Cowan et al. (1987). The range in age in these references is wide, between ~ 9 × 10⁹ and ~ 25 × 10⁹ yr depending on assumptions such as sudden synthesis or gradual enrichment. Stronger statements on narrower limits enhance the literature, but no one has denied that the problem is model dependent in a way that is difficult to check. A forceful review that favors the short end of this range can be enjoyed from Fowler (1987).

The astronomical age dating of the Galaxy comes from measured globular cluster main sequence termination points, combined with stellar evolutionary model calculations for various helium and metal chemical compositions. Age estimates became stabilized in the early 1980s near 17 × 10⁹ yr, based on generally accepted values of the chemical composition, the absolute magnitude M_v of the horizontal branches to set the termination luminosity, and models calculated by three principal groups (cf. Iben & Rood 1970, Ciardullo & Demarque 1977, VandenBerg 1983). The mean age of ~ 17 Gyr from the Yale (or VandenBerg) isochrones was based on distance moduli that use M_v = 0.63 mag for RR Lyrae stars of Osterhoff group II globular clusters and M_v = 0.80 mag for those of group I (Sandage 1982).

In a new development, observational evidence is becoming persuasive that the O/Fe chemical abundance ratio does not track with the Fe/H ratio in field subdwarfs of various Fe/H metallicities. An extensive review of the evidence is found in the ESO 1985 Workshop on Production and Distribution of C,N,O Elements, edited by Danziger et al. (1985). Following Simoda & Iben (1970), recent analyses by VandenBerg (1987), P. Demarque (private communication, 1987), and Rood & Crocker (1985) have independently shown that if the oxygen abundance remains high as the Fe/H ratio decreases, the age for a given turnoff luminosity decreases in the same way it would have decreased if Fe/H were increased. VandenBerg (1987) estimates that ages of globular clusters must be decreased by ~ 15% from the earlier values due to this effect of enhanced oxygen abundance in otherwise metal-poor stars if the effect in field subdwarfs occurs in globular cluster main sequence stars (but see Pilachowski 1988).

From this evidence and from a precision measurement of the age of 47 Tuc (Hesser et al. 1987), and adding 1.4 Gyr for the gestation time of the Galaxy, Sandage (1988a) adopted

for the age of the Universe from the age dating of stars.