We present a spectroscopic survey of the giant stellar stream found in the halo of the Andromeda galaxy .
Taken with the DEIMOS multi-object spectrograph on the Keck2 telescope , these data display a narrow velocity dispersion of 11 \pm 3 { km s ^ { -1 } } , with a steady radial velocity gradient of 245 { km s ^ { -1 } } over the 125 { kpc } radial extent of the stream studied so far .
This implies that the Andromeda galaxy possesses a substantial dark matter halo .
We fit the orbit of the stream in different galaxy potential models .
In a simple model with a composite bulge , disk and halo , where the halo follows a “ universal ” profile that is compressed by the formation of the baryonic components , we find that the kinematics of the stream require a total mass inside 125 { kpc } of M _ { 125 } = 7.5 ^ { +2.5 } _ { -1.3 } \times 10 ^ { 11 } { M _ { \odot } } , or M _ { 125 } > 5.4 \times 10 ^ { 11 } { M _ { \odot } } at the 99 % confidence level .
This is the first galaxy in which it has been possible to measure the halo mass distribution by such direct dynamical means over such a large distance range .
The resulting orbit shows that if M32 or NGC 205 are connected with the stream , they must either trail or lag the densest region of the stream by more than 100 { kpc } .
Furthermore , according to the best-fit orbit , the stream passes very close to M31 , causing its demise as a coherent structure and producing a fan of stars that will pollute the inner halo , thereby confusing efforts to measure the properties of genuine halo populations .
Our data show that several recently identified planetary nebulae , which have been proposed as evidence for the existence of a new companion of M31 , are likely members of the Andromeda Stream .