For what concerns me, Taxus baccata is one of the best woods for making bows, from prehistory to the Middle Ages. I know perfectly well that this is an obvious answer, but in my studies on the prehistoric weapons technology , analyses of this wood are very frequent. If you are interested, I have several reports linking the structure of the bow to the dendrological analysis (disposal the growth rings in the most stressed parts).
Apart from the suitability for making bows (see Beckhoff K (1968) Eignung und Verwendung einheimischer Holzarten für prähistorische Pfeilbogen. Die Kunde NF 19:85–101 [Suitability and use of native wood species for prehistoric bows]), it also plays a role that pollen of Taxus is notoriously difficult to identify and probably also poorly dispersed.
I did not realize that PDF's can be attached. Hereby the publication of Beckhoff (in German). His table 2 summarizes his results. Table 1 gives the scientific names to the German names used in table 2. "Eignung" in table 2 means suitability, and it ranges from very good (sehr gut) to bad (schlecht).
There is an overview article about finds of Dutch arrows published in 1999 by Lanting et al. Might be of interest for those involved in this subject. Find the PDF attached.
If you want other information on Taxus, I wrote a paper on it (it's in French, sorry!) because we have a lot of yew in the french Alps especially during Neolithic. It was also probably used in pastoral context.
I guess Otto is right : Taxus pollen can be well preserved if you have good pollen preservation (lake sediments for example) but in may cases it is very poorly preserved and underrepresented. Its recognition is requiring an experienced palynologist. There are many older pollen diagrams without Taxus curve which, ofcourse, should be there.....
In the Basque Mountains (Northern Spain) we have recently collected all Taxus palaeobotanical remains (pollen and charcoal) and we have seen a great use of yew wood in many archaeological sites from the Neolithic to Bronze age. At the same time we see low but significant presence in the pollen recods. The paper will be published soon.