> As to the remark on Drosera being closer to Nepenthes, I meant that
>they were close not only when comparing with other CPs, but other plants
>in general too. And don't take the 'close' so literally Jan. OK, there
>may be a 2 or 3 genera which go somewhere inbetween Drosera and Neps, but
>that still means they're relatively close, at least on that chart
>presented in that article.
OK, in this sense, they are really "relatively" close. But if you do not
swallow the caryophyllid affiliation (and do I have some problems here),
why do you believe the rest?
> Most of us do not know enough about plant systematics at high
>taxonomic ranks nor the relationships inferred from all the microscopic
>morphologic details (like polen and wood anatomy) to be able to compare the
>characteristics between dozens and dozens of plant families and agree
>that Neps are really not that far from Drosera.
Well, maybe, but this is quite exactly what a taxonomist *must* be able to
do. Else, macrosystematic statements are not well "rooted".
> I'm not saying they
>should or should not be in the same order, I'm just saying that for the
>CPers who grow these plants and are used to observing mainly the external
>macroscopic morphologic characteristics, these 2 genera do not appear to
>be closely related.
There are common traits like tetrad pollen and naphthoquinones, but
basically you are right. At least the classical Sarraceniales, including
Sarraceniaceae, Nepenthaceae, and Droseraceae are highly artificial and
certainly paraphyletic.
> Jan, I agree that phenotype must be considered in the 1st line.
(...)
>"monotheistic" classification based on a single DNA sequence probably won't
>live up to compete against one based on *numerous* phenotypic
>characteristics.
Well said, Fernando. I subscribe to this.
> But if we had a tree based on numerous DNA sequences
>and one based on numerous phenotypic characteristics, I would surely bet
>on the former.
Let's see. Until now I know only of one sequence which has been compared on
a rather large scale so it seems premature to make definitive statements.
> Again Jan, don't be so literal! This species was NOT on that chart
>and thruth is nobody knows anything about it! Nobody's ever grown the
>darn plant and few have ever even seen photos.
I have seen the plant (alive) and some photos, and I can tell you I believe
it is pretty close to _Drosophyllum_, this latter genus being a very very
weak Droseracea. So maybe it would be wise to exclude _Drosophyllum_
completely from Droseraceae. Further (DNA!) support to this views was
published recently in the 1995 supplement to Am.J.Bot. (the CAMERON thing).
_Triphyophyllum_ (Dioncophyllaceae) and a species of _Ancistrocladus_
(Ancistrocladaceae) were rbcL-sequenced and compared with the Droseraceae/
Nepenthaceae -clade. The result was that Dioncophyllaceae and
Ancistrocladaceae are most closely related to each other (OK, we have known
this since the famous palynological ERDTMAN paper, and from the fact that
they share naphthylisoquinoline alkaloids), another sister being...
_Drosophyllum_ (!!!). Now, an unresolved trichotomy joins the Droseraceae
line (excl. _Drosophyllum_), the Nepenthaceae line, and the
Dioncophyllaceae/ Ancistrocladaceae/ _Drosophyllum_ line. This is a further
indication for paraphyly of Droseraceae in the traditional sense (incl.
_Drosophyllum_). Veterans on this discussion list may remember that I have
supposed this some while ago (on purely morphological grounds), already.
You see how addition of few data can modify the tree published originally.
I suppose some further modification will follow in due course (maybe even
leading to the exclusion of the cps from the caryophyllid group, which
would rather dramatically change -and I think improve- the information of
the tree). I think this case rather aptly demonstrates the potential of DNA
sequence comparison (i.e. being not too strong in a priori statements but
rather suitable for mutual corroboration of "traditional" and molecular
data).
>> >Yeah, it isn't hard to reach that conclusion without the DNA testing
>> >but wouldn't this lend support to the validity of this type of comparison?
>>
>> I would formulate it the other way round: this type of comparison lends
>> support to the conclusions from comparative morphology. All other
>> implications are at least dangerous.
>
> I would say it another way: wouldn't both support each other?!?
If you insist on this formulation, I could perhaps agree.
>(...)Don't you agree that genes show and record these pathways better?
Only if you have several genes, or even better, all genes. How about an all
plant taxa genome project? ;-)
This would keep some people busy for some time but it rather certainly
would eventually lead to a nearly perfect phylogenetic reconstruction.
So long ;-)
Kind regards
Jan