We present a survey of helium abundance indicators derived from a comprehen
sive study of globular cluster photometry in the literature. For each of th
e three indicators used, we conduct a thorough error analysis, and identify
systematic errors in the computational procedures. For the population rati
o R=N-HB/N-RGB, we find that there is no evidence of a trend with metallici
ty, although there appears to be real scatter in the values derived. Althou
gh this indicator is the one best able to provide useful absolute helium ab
undances, the mean value is Y approximate to 0.20, indicating the probable
presence of additional systematic error.
For the magnitude difference from the horizontal branch to the main sequenc
e Delta and the RR Lyrae mass-luminosity exponent A, it is only possible to
determine relative helium abundances reliably. This is due to continuing u
ncertainties in the absolute metallicity scale for Delta, and uncertainty i
n the RR Lyrae temperature scale for A. Both indicators imply that the heli
um abundance is approximately constant as a function of [Fe/H]. According t
o the A indicator, both Oosterhoff I and II group clusters have constant va
lues independent of [Fe/H] and horizontal branch type. In addition, the two
groups have slopes d log[P-ab]/d[Fe/H] that are consistent with each other
, but significantly smaller than the slope for the combined sample.