What is correct: POWO page listed name or specimen name in herbarium sheets ?:
I have been encountering this issue both in POWO and GBIF.

The Listed name (including synonyms mentioned) is different on the page (e.g. in POWO and specimen details, it is given as Lectotype of Mussaenda macrophylla Wall.), but many times I find the name given in the herbarium sheet (Mussaenda treutleri here) is different.
So what should we consider it: Mussaenda macrophylla Wall. or Mussaenda treutleri ?

You are missing something.

Mussaenda frondosa var. grandifolia Hook.f. = Mussaenda treutleri Stapf.
Mussaenda macrophylla var. grandisepala (Jayaw.) Alejandro = Mussaenda macrophylla Wall.

But in the herbarium sheet, it is mentioned as Mussaenda frondosa var. grandifolia Hook.f., which is a syn. of  Mussaenda treutleri Stapf. as per POWO.

But it is listed as a Lectotype of Mussaenda macrophylla Wall.

It could just be a mistake. May be Jayweera made a mistake and designated a wrong lectotype.

My question is ‘What is correct: POWO page listed name or specimen name in herbarium sheets ?’ as a layman.

This has happened to me many times with me, with both POWO and GBIF.

On closely looking at the sepals, bracteoles etc., Jayweera’s this specimen is more closer to his other specimen of Mussaenda treutleri Stapf. only as per specimen listed at POWO

To keep ourself safe, we should look at accession Number.

The specimen cited . macrophylla Wall is the holotype, meaning a single specimen selected by the original author. This has accession No. K000031997;
The specimen cited with M. treutleri Stapf, distinct accepted species name (syn: M. frondosa var. grandifolia Hook.f.), specimen cited is a syntype (one of several specimens cited together by author) with accession number K000031996
The specimen cited by … is Lectotype (one selected from several syntypes) of M. trutleri Stapf, with accession number K000031995

But what is the answer to my question as a layman?

Wallich cited several specimens in his publication Wallich Catalogue 1812 (as there were no descriptions publication and names remained illegitimate unless validated by some later author/s). Mussaenda macrophylla Wall. was validated in Roxb.Fl. Ind. 1824, based on the Wall. Catalogue No.. 6295. Hooker also included M. frondosa Wall. Cat. 6250 A under it. M. frondosa 6250 E he separated under frondosa var. grandifolia Hook.f.. Jayaweera 1960-61 who examined these specimens realised that Wall. Cat. 6295 is true M. macrophylla Wall. designated it as Holotype (although it is not technically right), and shifted M. frondosa var. grandifolia Hook.f as distinct from M. frondosa Linn. as M. treutleri Stapf, as the name M. grandifolia Elmer was preoccupied.  and selected one specimen as Lectotype out of several syntypes. Wall. Cat. 6250 had several specimens under numbering A, B. C, D, E, F, G etc,, which went different ways.

as I have mentioned earlier also, annotations on the herbarium sheets are made by workers who visit the herbarium physically and note their observations. Many researchers may look at the specimens virtually and make their conclusions, which may or may not be globally accepted. I have very clearly explained how M. macrophylla Wall. and M. treutleri Stapf.

Should we generally go by the name which is written in the herbarium sheet rather than those depicted in the POWO/ GBIF pages, in case of such variations?

Your answer is not for a layman.
My question is:
POWO/ GBIF are overriding what is written in the herbarium sheets (without changing its name on it) in such cases. Should we go by the opinion of POWO/GBIF as shown in their databases or Should we generally go by the name which is written in the herbarium sheet ? Or it can be either way ? Or do we compare with specimens where there is no such ambiguity ?
I require this guidance as I am very frequently trying to match with POWO/ GBIF specimens and I find such conflicts many times.

I would prefer to use the POWO/GBIF database name rather than the name found on the herbarium sheet which may not be up-to-date. As … pointed out, the specimens especially Holotypes would/should have the latest name reflecting the database name, but other specimens may still have old names on them. Moreover, I would assume, the scanned images of the specimens that we see may be old and not reflect any recent changes in the nomenclature. So, we should go by the database name which is supposed to be the most recent accepted name.

There is another problem here. Although I also like to attach determinavits, I do it only when I am sure. But then people do attach slips mistakenly and this wrong ID sometimes persists.

Secondly, one collection can have multiple specimens on the sheets or multiple specimens on multiple sheets. Someone while revising may describe a new species based on this collection of supposedly same species.
1. A normal person doesn’t go and check the types.
2. It all depends on your opinion why you may follow a website’s name or a name on the sheet. Just for example, I described Gastrochilus kadooriei from fresh as well as a collection of plants which were deposited as G. pseudodistichus. So now on the sheet you will see the name G. pseudodistichus but the correct name is G. kadooriei. To help with this issue, I went to check different collections at Kunming and Kew and I changed the names. I requested my friends in Vietnam, Laos and Russia to change the names accordingly and they did. While Kunming didnt update the names on their website, others did. But my new species is 100% accepted.
When Garg sir raised this question, it was more out of eagerness to learn and it shows that he is more than a citizen scientist. While I have seen many scientists, who never check the types.
On the other hand, super experts like … are also correct in their explanation and so is …
There are many reasons why curators do not like to write anything on the holotype sheets. Some of these sheets are more than 250 years old. It is like an antique and heritage and it needs to be preserved properly for lifetime. It needs a hell lot of money to keep these specimens safe. At many places like Kew, a normal person would not even get access to these sheets. I was lucky in that sense, wherever I went I was allowed to check them. You will be surprised, I heard that one top organisation banned entry of a top scientists to their herbarium, because this scientist damaged some of the type specimens.

I hope I have more clarity on this now.

Great explanation, …! That’s so meticulous of you to update all specimens. But as you said it’s not always practical.
Anyone who is capable of damaging such rare Type specimens due to carelessness, should not be allowed in any Herbarium, in my opinion.