We present a new analysis of the properties of the young massive star clusters forming profusely in intense starburst environments , which demonstrates that these objects are plausible progenitors of the old globular clusters ( GCs ) seen abundantly in the Local Group .
The method is based on the tight relationship for old GCs between their V -band luminosities , L _ { V } , and ( central ) velocity dispersions , \sigma _ { 0 } .
We improve the significance of the relationship by increasing the GC sample size and find that its functional form , L _ { V } / L _ { \odot } \propto \sigma _ { 0 } ^ { 1.57 \pm 0.10 } ( km s ^ { -1 } ) , is fully consistent with previous determinations for smaller Galactic and M31 GC samples .
The tightness of the relationship for a GC sample drawn from environments as diverse as those found in the Local Group implies that its origin must be sought in intrinsic properties of the GC formation process itself . We evolve the luminosities of those young massive star clusters ( YMCs ) in the local Universe which have velocity dispersion measurements to an age of 12 Gyr , adopting a variety of IMF descriptions , and find that most YMCs will evolve to loci close to , or to slightly fainter luminosities than the improved GC relationship .
In the absence of significant external disturbances , this implies that these objects may potentially survive to become old GC-type objects over a Hubble time .
The main advantage of our new method is its simplicity .
Where alternative methods , based on dynamical mass estimates , require one to obtain accurate size estimates and to make further assumptions , the only observables required here are the system ’ s velocity dispersion and luminosity .
The most important factor affecting the robustness of our conclusions is the adopted form of the initial mass function .
We use the results of N -body simulations to confirm that dynamical evolution of the clusters does not significantly alter our conclusions about the likelihood of individual clusters surviving to late times .
Finally , we find that our youngest observed clusters are consistent with having evolved from a relation of the form L _ { V } / L _ { \odot } \propto \sigma _ { 0 } ^ { 2.1 _ { -0.4 } ^ { +0.5 } } ( km s ^ { -1 } ) .
This relation may actually correspond to the origin of the GC fundamental plane .