That the mitochondrial mutation rate (set at about 1 mutation in 600 generations) is likely much higher than assumed by Sykes can be seen at the site of The Molecular History Research Center at
Some studies have shown a mutation rate of 1 in about 40 generations, see e.g. http://www.ridgenet.net/~do_while/sage/v2i6n.htm
Sykes places Ursula at 40000 BC in Greece whereas the oldest archaeological records of first humans in Greece would seem to be about 20000 BC as at Franchthi, Greece - see http://emuseum.mnsu.edu/archaeology/sites/europe/franchthicave.html
so that a good argument could be made that all of Sykes dates should at least be halved.
If one does so then the dates correspond better with the dating of archaeological finds in the respective regions.
Xenia at [Sykes' calculated] 24000 years ago would [then more correctly] be ca. 12000 years ago, a date also assigned by the linguists to the spread of Indo-European from this central European region and close to the Mesolithic skulls found dating to ca. 8000 BC.
Helena in France [dated by Sykes] at 20000 years ago would [then more correctly] be ca. 10000 years ago and thus near the date of 9000 BC I assign to the astronomical paintings at Lascaux.
Velda in Spain [as dated by Sykes] at 17000 years would [then more correctly] be in the same chronologicalball park at ca. 9000 BC for the Altamira cave paintings near Santander.´
So as one can see, there is still much that we need to know and learn.