> Now, another question more interesting for our scientists friends:  Adrian
> Slacks, in his book, talks about Drosera erithroriza (wrong spelling maybe)
> like a non-carnivorous plant.  The leaves drag no tentacles, so they can
> not catch insects (logical !) but however, they grow very well.
> 
> That's ok,
No. That's wrong. Some specimens have reduced galndular zones on the 
laminae, sometimes extremely so. However, usually _D. erythrorhiza_ 
has normal trapping leaves.
> but in another book (Gerard Blondeau) I see a marvellous picture
> of a plant called Drosera erithroriza with short, but really red and
> numerous tentacles.
I don't know the book but the lable could be correct.
> Are there a lot of sub-species in this family ?
The family = Droseraceae (3 genera, some 137+ species, quite a lot of 
subspecies). The (formally acknowledged) subspecies are all found in 
the genus _Drosera_.  _Drosera erythrorhiza_ has the subspp. 
_collina, erythrorhiza, magna_, and _squamosa_, in addition to two 
unaccepted varieties (one of them is based on the basionym of subsp. 
_squamosa_, however) and one likewise unaccepted form (invalid and 
based on the other var. imbecilia).
> Or is one of the two auctors wrong ?
Yes, Slack (p.p., if he really wrote what you have cited).
> If anyone could answer me this time...
If you accept this reply as an answer. I do unfortunately not have 
any own observations on your _Nepenthes_ water level problem. I could 
design a simple experimental setup for "doing it yourself" but I trust 
you will have the same ideas.
Kind regards
Jan