Originally it was thought that all globular clusters
were part of the halo. Now, however, it is realized that two
distinct populations of globulars exist. Old, metal-poor clusters
([Fe/H] < -0.8) are part of an extended, spherical halo, while
younger clusters with [Fe/H] > -0.8 are in a more concentrated and
flattened distribution.
The less-metal-poor clusters have a scale height similar to that of the thick disk, and they may be associated. Other ideas have them related to the Galaxy's bulge instead..
Globular Cluster ages:
Young: 9-12 billion years oldField Stars
Old: 12-14 billion years old
but there's still a lot of controversy about absolute ages here...
How do we find field halo stars? Look at the space velocities of stars with respect to the Sun. If they are low, they are probably disk stars (like the Sun). If they are high, they are usually associated with the halo.If we add up all the mass in the field stars and the metal-poor GCs, we can come up with a rough density distribution for the Galaxy's stellar halo:Like the GCs, the halo field stars are also metal-poor. This tells us something about the formation of the Galaxy!
The total mass of the halo is about 109 M
sun, about 1% of which is the globular clusters, and the rest in
field stars. So there's not a lot of stuff in the stellar halo, but
what is there holds a lot of information about the early history of the
Galaxy!
Below: Star streams in the galactic halo from the SDSS survey
(Belokurov et al 2006). Upper main sequence/turnoff stars have been
selected via a color cut; these stars should have similar luminosities,
so their apparent magnitude is a measure of distance. In this plot,
color is not the color of the star, but distance (blue nearer, red
further).

On larger scales, these streams can be overlaid on 2MASS
measurements of structure in the galactic halo, showing how the big
SDSS stream connects with the larger Sagittarius stream detected by
2MASS:
